A product quality analysis method and system based on industrial big data

Abstract

The present invention relates to the technical field of product analysis, and discloses a product quality analysis method and system based on industrial big data, comprising obtaining a first comparison result based on initial quality data and standard quality data, obtaining a second comparison result based on a real-time surface image and a standard surface image, inputting a first vector and a second vector into a pre-constructed processing improvement model, obtaining initial honey processing data, correcting the initial honey processing data according to processing equipment data, obtaining target honey processing data, and finally obtaining target quality data of processed honey, and performing analysis according to the target quality data; the present invention first performs quality analysis on unprocessed honey, thereby determining subsequent processing parameters according to the quality data of the unprocessed honey, and finally performs quality analysis on the processed honey, so that not only the processing effect of honey is improved, but also more accurate honey quality analysis results can be obtained.

Description

Technical Field

[0001] The present invention relates to the field of product analysis technology, and more specifically, to a product quality analysis method and system based on industrial big data. Background Art

[0002] Industrial big data refers to data sets generated and collected in the industrial field. These data are processed through various advanced analytical technologies and tools to improve production efficiency, optimize processes, predict maintenance needs, reduce costs and improve decision-making levels. Now, industrial big data has been applied to many fields, especially in product quality inspection. Through data collection, analysis and processing, it can significantly improve quality inspection efficiency, accuracy and reliability. In the existing technology, data analysis technology has been used in the quality assessment of honey, but there are still certain problems.

[0003] For example, Chinese patent publication number CN102313715A provides a honey quality detection method based on laser technology, which collects the near-infrared spectrum of the honey to be tested, pre-processes the near-infrared spectrum, and uses a genetic algorithm to analyze the pre-processed near-infrared spectrum, paving the way for subsequent honey quality analysis; Chinese patent publication number CN110244009A provides a rapid identification and detection system and method for honey quality, which directly uses a laser to irradiate the honey to be tested, and then receives the laser spot image formed by it for analysis, and analyzes the quality of the honey based on the laser spot image.

[0004] Although the above patents all use data analysis technology to analyze the quality of honey, they usually perform quality analysis on processed honey, but ignore the state of honey before processing. For example, in the existing honey processing process, the quality of honey crystallization is improved through the microbubble process, and the microbubble process parameters are related to the state of honey before processing. Therefore, in the process of honey quality analysis, comprehensive considerations should be made to obtain accurate quality analysis results.

[0005] In view of this, the present invention proposes a product quality analysis method and system based on industrial big data to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a product quality analysis method and system based on industrial big data.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A product quality analysis method based on industrial big data, comprising:

[0009] S10: acquiring initial quality data and a real-time surface image of unprocessed honey, obtaining a first comparison result based on the initial quality data and the standard quality data, and obtaining a second comparison result based on the real-time surface image and the standard surface image;

[0010] S20: determining a corresponding first weight based on the first comparison result, determining a corresponding second weight based on the first weight, determining a first vector based on the first comparison result and the first weight, and determining a second vector based on the second comparison result and the second weight;

[0011] S30: inputting the first vector and the second vector into a pre-built processing improvement model to obtain initial honey processing data;

[0012] S40: Acquire processing equipment data, and correct the initial honey processing data according to the processing equipment data to obtain target honey processing data, wherein the processing equipment is equipment required for processing honey;

[0013] S50: Obtain target quality data of processed honey, perform analysis according to the target quality data, and obtain analysis results, wherein the processed honey is obtained by processing unprocessed honey according to the target honey processing data.

[0014] Furthermore, the method for obtaining a second comparison result based on the real-time surface image and the standard surface image includes:

[0015] The real-time surface image and the standard surface image are binarized respectively, and a first shadow area corresponding to the real-time surface image and a second shadow area corresponding to the standard surface image are calculated, and a ratio of the first shadow area to the second shadow area is taken as a second comparison result.

[0016] Furthermore, the method for determining the corresponding first weight based on the first comparison result includes:

[0017]

[0018] In the formula, PRW is the first weight, Q rj is the initial value of the jth element in the first comparison result, Q sj is the standard value of the jth element in the first comparison result, n is the number of elements in the first comparison result, tan -1 (·) is the inverse tangent function, and π is a constant.

[0019] Furthermore, the method for determining the first vector according to the first comparison result and the first weight includes:

[0020] The first comparison result is converted into a first comparison vector, the first weight is converted into a first weight vector, and the first comparison vector and the first weight vector are cascaded and fused to generate a first vector.

[0021] Furthermore, the method for constructing the processing improvement model includes:

[0022] A processing sample data set is obtained, wherein the processing sample data set includes a first historical vector, a second historical vector and historical honey processing data, the processing sample data set is divided into a sample training set and a sample test set, a regression network is constructed, the first historical vector and the second historical vector in the sample training set are used as input data of the regression network, the historical honey processing data in the sample training set are used as output data of the regression network, the regression network is trained to obtain an initial regression network for predicting real-time honey processing data, the initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as a processing improvement model.

[0023] Furthermore, the method for correcting the initial honey processing data according to the processing equipment data includes:

[0024] The equipment aging coefficient is calculated based on the processing equipment data, and the initial honey processing data and the equipment aging coefficient are input into the aging correction model to obtain the target honey processing data.

[0025] Furthermore, the processing equipment data includes the number of maintenance records, the total operating time and the startup time. The method for calculating the equipment aging coefficient based on the processing equipment data includes:

[0026]

[0027] EAC is the equipment aging coefficient, MRC is the number of maintenance records, TOE is the total operating time, STE is the start time, log 2 (·) is the logarithmic function with base 2, and cosh(·) represents the hyperbolic cosine function.

[0028] Furthermore, the method for constructing the aging correction model includes:

[0029] An aging sample data set is obtained, wherein the aging sample data set includes historical initial honey processing data, historical equipment aging coefficients, and historical target honey processing data. The aging sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical initial honey processing data and the historical equipment aging coefficients in the sample training set are used as input data of the regression network, and the historical target honey processing data in the sample training set are used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting real-time target honey processing data. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as an aging correction model.

[0030] Furthermore, the standard quality data includes a standard fructose ratio, a standard viscosity value, a standard water content and a standard hydroxymethylfurfural concentration value, and the method for obtaining the first comparison result based on the initial quality data and the standard quality data includes:

[0031] The initial mass data and the standard mass data are traversed, the differences between the corresponding elements are calculated, a mass difference sequence is generated, and the mass difference sequence is used as the first comparison result. The mass difference sequence includes the difference in fructose ratio, the difference in viscosity, the difference in water content and the difference in hydroxymethylfurfural concentration.

[0032] A product quality analysis system based on industrial big data, which is used to implement the above-mentioned product quality analysis method based on industrial big data, comprises:

[0033] A comparison generation module: used for obtaining initial quality data and a real-time surface image of unprocessed honey, obtaining a first comparison result based on the initial quality data and the standard quality data, and obtaining a second comparison result based on the real-time surface image and the standard surface image;

[0034] A weight generation module: used to determine a corresponding first weight based on a first comparison result, determine a corresponding second weight based on the first weight, determine a first vector according to the first comparison result and the first weight, and determine a second vector according to the second comparison result and the second weight;

[0035] Improvement module: used for inputting the first vector and the second vector into the pre-built processing improvement model to obtain initial honey processing data;

[0036] Correction module: used to obtain processing equipment data, correct the initial honey processing data according to the processing equipment data, and obtain target honey processing data, wherein the processing equipment is the equipment required for processing honey;

[0037] Analysis module: used to obtain target quality data of processed honey, analyze according to the target quality data, and obtain analysis results. The processed honey is obtained by processing unprocessed honey according to the target honey processing data.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention first obtains a first comparison result based on initial quality data and standard quality data, obtains a second comparison result based on a real-time surface image and a standard surface image, then determines a corresponding first weight based on the first comparison result, determines a corresponding second weight based on the first weight, inputs the first vector and the second vector into a pre-constructed processing improvement model, obtains initial honey processing data, obtains processing equipment data, corrects the initial honey processing data according to the processing equipment data, obtains target honey processing data, and finally obtains target quality data of processed honey, performs analysis according to the target quality data, and obtains analysis results. The present invention first performs quality analysis on unprocessed honey, thereby determining subsequent processing parameters according to the quality data of the unprocessed honey, and finally performs quality analysis on the processed honey, which not only improves the processing effect of honey, but also can obtain more accurate honey quality analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a flow chart of a product quality analysis method based on industrial big data in the present invention;

[0041] Figure 2 It is a structural schematic diagram of a product quality analysis system based on industrial big data in the present invention;

[0042] Figure 3 A schematic diagram of a computer-readable storage medium in the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] Example 1

[0045] See also Figure 1 As shown, this embodiment discloses a product quality analysis method based on industrial big data, including:

[0046] S10: Acquire initial quality data and a real-time surface image of unprocessed honey, obtain a first comparison result based on the initial quality data and the standard quality data, and obtain a second comparison result based on the real-time surface image and the standard surface image.

[0047] It should be noted that unprocessed honey is the honey obtained by filtering after being transported from the honey producing area to the processing area. Filtration only removes impurities such as honeycomb fragments, bee and insect remains in the honey, and does not change the quality of the honey itself. The initial quality data includes the initial fructose-glucose ratio, initial viscosity value, initial water content and initial hydroxymethylfurfural concentration value. The fructose-glucose ratio refers to the ratio of glucose content to fructose content in honey. Hydroxymethylfurfural is one of the important indicators for evaluating the quality and freshness of honey. The above-mentioned glucose content, fructose content and hydroxymethylfurfural concentration values ​​are obtained by testing honey samples by high performance liquid chromatography, the viscosity value is obtained by testing honey samples by capillary viscometer method, and the water content is obtained by testing honey samples by refractometer method based on the principle of refraction of light in media of different densities.

[0048] The standard quality data includes a standard fruit-to-glucose ratio, a standard viscosity value, a standard water content, and a standard hydroxymethylfurfural concentration value. The method for obtaining the first comparison result based on the initial quality data and the standard quality data includes:

[0049] The initial mass data and the standard mass data are traversed, the differences between the corresponding elements are calculated, a mass difference sequence is generated, and the mass difference sequence is used as the first comparison result. The mass difference sequence includes the difference in fructose ratio, the difference in viscosity, the difference in water content and the difference in hydroxymethylfurfural concentration.

[0050] It is understandable that the above-mentioned standard quality data is preset based on expert experience and stored in a database so as to be accessible in real time.

[0051] The method for obtaining a second comparison result based on the real-time surface image and the standard surface image includes:

[0052] The real-time surface image and the standard surface image are binarized respectively, and a first shadow area corresponding to the real-time surface image and a second shadow area corresponding to the standard surface image are calculated, and a ratio of the first shadow area to the second shadow area is taken as a second comparison result.

[0053] It should be added that the shaded area above represents the impurities that have not been filtered out of the honey, or the surface texture of the honey, etc. The standards for binarizing the real-time surface image and the standard surface image are the same. The first shadow area and the second shadow area are obtained by using the 0penCV counting non-zero pixel function to calculate the total number of pixels in the shadow (black) area in the binary image. The second comparison result represents the difference between the real-time surface image and the standard surface image. The closer the second comparison result is to 1, the better the quality of the honey.

[0054] S20: Determine a corresponding first weight based on the first comparison result, determine a corresponding second weight based on the first weight, determine a first vector according to the first comparison result and the first weight, and determine a second vector according to the second comparison result and the second weight.

[0055] The method for determining the corresponding first weight based on the first comparison result includes:

[0056]

[0057] In the formula, PRW is the first weight, Q rj is the initial value of the jth element in the first comparison result, Q sj is the standard value of the jth element in the first comparison result, n is the number of elements in the first comparison result, tan -1 (·) is the inverse tangent function, and π is a constant.

[0058] For example, if the first comparison result includes the difference in the fruit-glucose ratio, the difference in the viscosity value, the difference in the water content, and the difference in the hydroxymethylfurfural concentration, then n is 4, Q rj -Q sj is one of the above differences.

[0059] The method for determining a corresponding second weight based on a first weight includes:

[0060] SEW = 1-PRW;

[0061] Where SEW is the second weight.

[0062] The method for determining the first vector according to the first comparison result and the first weight includes:

[0063] The first comparison result is converted into a first comparison vector, the first weight is converted into a first weight vector, and the first comparison vector and the first weight vector are cascaded and fused to generate a first vector.

[0064] It can be understood that cascade fusion refers to concatenating two vectors end to end to generate a new vector, that is, connecting the two vectors in sequence into a longer vector, and converting the first comparison result and the first weight into a vector through the One-Hot Encoding method. The above cascade fusion is only an example, and this embodiment is not limited to this.

[0065] The method for determining the second vector according to the second comparison result and the second weight includes:

[0066] The second comparison result is converted into a second comparison vector, the second weight is converted into a second weight vector, and the second comparison vector and the second weight vector are cascaded and fused to generate a second vector.

[0067] Similarly, the implementation method is the same as the above content, and this embodiment will not go into details about it.

[0068] S30: Inputting the first vector and the second vector into a pre-built processing improvement model to obtain initial honey processing data.

[0069] In this embodiment, the initial honey processing data includes but is not limited to stirring time, heating temperature and pressurization time. It can be understood that the first comparison result includes the difference in fruit-to-glucose ratio, viscosity value difference, water content difference and hydroxymethylfurfural concentration difference.

[0070] The above-mentioned difference in the fructose ratio is related to the pressurization time. The fructose ratio refers to the ratio of the glucose content to the fructose content in honey. The crystallization of honey is also related to the glucose and fructose contents. At the same temperature, the solubility of glucose is much smaller than that of fructose, and the soluble glucose content is higher, and the honey crystallizes faster. Therefore, in this embodiment, the larger the difference in the fructose ratio, the higher the soluble glucose content of the unprocessed honey, and therefore the shorter the required pressurization time, because the pressurization time refers to the time for passing gas into the honey during the stirring process of the honey, and the purpose of pressurization is to make the honey crystallization effect better; it is understandable that those skilled in the art can set the standard fructose ratio to a lower value, then the initial fructose ratio is usually greater than the standard fructose ratio, and the fructose ratio difference is the initial fructose ratio minus the standard fructose ratio, so that the fructose ratio difference can characterize the soluble glucose content of the unprocessed honey.

[0071] The above-mentioned difference in viscosity values ​​is related to the stirring time. It can be understood that the larger the difference in viscosity values, the greater the viscosity of the unprocessed honey, and therefore the shorter the stirring time required.

[0072] The above-mentioned water content difference and hydroxymethylfurfural concentration difference are related to the heating temperature. The larger the water content difference, the higher the water content of the unprocessed honey, and the higher the required heating temperature; the larger the hydroxymethylfurfural concentration difference, the higher the hydroxymethylfurfural concentration of the unprocessed honey, and the lower the required heating temperature, because a higher heating temperature will increase the hydroxymethylfurfural concentration in the honey.

[0073] The second comparison result also affects the initial honey processing data. The closer the second comparison result is to 1, the better the quality of the honey, which means that the required stirring time and pressurizing time are shorter, and the required heating temperature is lower. As can be seen from the above content, the second comparison result is the ratio of the first shadow area to the second shadow area. The shadow area represents the impurities that are not filtered out of the honey, or the surface texture of the honey, etc. Similarly, if a standard surface image with a smaller second shadow area is pre-selected, the second comparison result will usually be greater than 1. The larger the second comparison result, the more unfiltered impurities in the surface honey or the more surface texture of the honey, which requires longer stirring time and pressurizing time, as well as higher heating temperature, because stirring and pressurizing are conducive to the floating of impurities, and higher heating temperature separates the fine particles in the honey from the impurities.

[0074] The construction method of the processing improvement model includes:

[0075] A processing sample data set is obtained, wherein the processing sample data set includes a first historical vector, a second historical vector and historical honey processing data. The processing sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The first historical vector and the second historical vector in the sample training set are used as input data of the regression network, and the historical honey processing data in the sample training set are used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting real-time honey processing data. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as a processing improvement model. The initial regression network is preferably a deep neural network model.

[0076] It should be added that the above-mentioned processing improvement model receives the first vector and the second vector, and can perform reverse decoding to obtain the corresponding first comparison result and the second comparison result, and determine the corresponding numerical intervals according to the mapping relationship. Taking the fruit-to-grape ratio difference and the viscosity value difference in the first comparison result as an example, the numerical interval corresponding to the fruit-to-grape ratio difference is determined according to the mapping relationship, and then the corresponding pressurization time is determined according to the numerical interval. The numerical interval corresponding to the viscosity value difference is determined according to the mapping relationship, and then the corresponding stirring time is determined according to the numerical interval.

[0077] It can be understood that the pressurization time and stirring time in the initial honey processing data are used as examples to illustrate that the pressurization time is affected by the difference in the fruit-to-grape ratio in the first comparison result, and is also affected by the second comparison result, and the first comparison result and the second comparison result have corresponding first weights and second weights, respectively. Similarly, the stirring time is affected by the difference in the viscosity values ​​in the first comparison result, and is also affected by the second comparison result.

[0078] It can be understood that the mapping relationship mentioned above is pre-established based on expert experience or experiments conducted by technical personnel in this field. By establishing a mapping relationship between the first historical vector, the second historical vector and the historical honey processing data, the model is trained to obtain a processing improvement model.

[0079] S40: Acquire processing equipment data, and correct the initial honey processing data according to the processing equipment data to obtain target honey processing data, wherein the processing equipment is the equipment required for processing honey.

[0080] It can be seen from the above content that the initial honey processing data includes but is not limited to stirring time, heating temperature and pressurizing time. In this embodiment, the processing equipment may be a high-pressure reactor, which can perform stirring, heating and pressurizing.

[0081] In this embodiment, the processing equipment data includes but is not limited to the number of maintenance records, the total operating time, the start-up time, and the start-up time. The start-up time is the sum of the stirring start-up time, the heating start-up time, and the pressurization start-up time. In this embodiment, the start-up time refers to the time required for the processing equipment to go from the starting working state to the stable working state. Taking the stirring start-up time as an example, the stirring start-up time refers to the time required from the start of rotation of the stirring shaft to the target speed. It can be understood that the above-mentioned start-up time is recorded by a sensor; still taking the stirring start-up time as an example, by setting a sensor on the stirring shaft, the start time is recorded from the beginning of rotation of the stirring shaft until the stirring shaft reaches the target speed, thereby ending the recording.

[0082] Methods for correcting initial honey processing data based on processing equipment data include:

[0083] The equipment aging coefficient is calculated based on the processing equipment data, and the initial honey processing data and the equipment aging coefficient are input into the aging correction model to obtain the target honey processing data.

[0084] Methods for calculating equipment aging coefficient based on processing equipment data include:

[0085]

[0086] EAC is the equipment aging coefficient, MRC is the number of maintenance records, TOE is the total operating time, STE is the start time, log 2 (·) is the logarithmic function with base 2, and cosh(·) represents the hyperbolic cosine function.

[0087] In this embodiment, the number of maintenance records, the total operating time, and the startup time are all positively correlated with the aging degree of the processing equipment. Therefore, it can be seen from the above formula that the equipment aging coefficient is positively correlated with the aging degree of the processing equipment.

[0088] The method for constructing the aging correction model includes:

[0089] An aging sample data set is obtained, wherein the aging sample data set includes historical initial honey processing data, historical equipment aging coefficients, and historical target honey processing data. The aging sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical initial honey processing data and the historical equipment aging coefficients in the sample training set are used as input data of the regression network, and the historical target honey processing data in the sample training set are used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting real-time target honey processing data. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as an aging correction model. The initial regression network is preferably a recurrent neural network model.

[0090] It is understandable that, according to the same principle, when the aging correction model receives the initial honey processing data and the equipment aging coefficient, it can first determine the coefficient interval to which the equipment aging coefficient belongs, and then increase the initial honey processing data according to the increase corresponding to the coefficient interval, so as to obtain the target honey processing data. It is understandable that, under normal circumstances, due to the aging of the processing equipment, the processing equipment cannot accurately process the honey according to the initial honey processing data, and therefore the initial honey processing data needs to be improved accordingly to ensure the honey processing effect.

[0091] S50: Obtain target quality data of processed honey, perform analysis according to the target quality data, and obtain analysis results, wherein the processed honey is obtained by processing unprocessed honey according to the target honey processing data.

[0092] In this embodiment, the target quality data includes but is not limited to a target hydroxymethylfurfural concentration value, a target viscosity value, a target water content and a target conductivity. The target conductivity represents the mineral content in the honey. The larger the target conductivity, the more mineral content in the honey and the better the quality of the honey.

[0093] It should be noted that the analysis results include superior honey, first-grade honey and second-grade honey. Analysis based on target quality data means comparing the target quality data with the Chinese honey standard GB14963-2011, or with the international honey standard to obtain the analysis results.

[0094] In this embodiment, a first comparison result is first obtained based on the initial quality data and the standard quality data, a second comparison result is obtained based on the real-time surface image and the standard surface image, a corresponding first weight is determined based on the first comparison result, a corresponding second weight is determined based on the first weight, the first vector and the second vector are input into a pre-constructed processing improvement model, initial honey processing data is obtained, processing equipment data is obtained, the initial honey processing data is corrected according to the processing equipment data, target honey processing data is obtained, and finally target quality data of the processed honey is obtained, and analysis is performed according to the target quality data to obtain analysis results; in this embodiment, quality analysis is first performed on the unprocessed honey, so as to determine subsequent processing parameters according to the quality data of the unprocessed honey, and finally the processed honey is subjected to quality analysis, which not only improves the processing effect of honey, but also can obtain more accurate honey quality analysis results.

[0095] Example 2

[0096] like Figure 2 As shown, this embodiment provides a product quality analysis system based on industrial big data on the basis of embodiment 1, including:

[0097] Comparison generation module: used to obtain initial quality data and real-time surface image of unprocessed honey, obtain a first comparison result based on the initial quality data and standard quality data, and obtain a second comparison result based on the real-time surface image and the standard surface image.

[0098] It should be noted that unprocessed honey is the honey obtained by filtering after being transported from the honey producing area to the processing area. Filtration only removes impurities such as honeycomb fragments, bee and insect remains in the honey, and does not change the quality of the honey itself. The initial quality data includes the initial fructose-glucose ratio, initial viscosity value, initial water content and initial hydroxymethylfurfural concentration value. The fructose-glucose ratio refers to the ratio of glucose content to fructose content in honey. Hydroxymethylfurfural is one of the important indicators for evaluating the quality and freshness of honey. The above-mentioned glucose content, fructose content and hydroxymethylfurfural concentration values ​​are obtained by testing honey samples by high performance liquid chromatography, the viscosity value is obtained by testing honey samples by capillary viscometer method, and the water content is obtained by testing honey samples by refractometer method based on the principle of refraction of light in media of different densities.

[0099] The standard quality data includes a standard fruit-to-glucose ratio, a standard viscosity value, a standard water content, and a standard hydroxymethylfurfural concentration value. The method for obtaining the first comparison result based on the initial quality data and the standard quality data includes:

[0100] The initial mass data and the standard mass data are traversed, the differences between the corresponding elements are calculated, a mass difference sequence is generated, and the mass difference sequence is used as the first comparison result. The mass difference sequence includes the difference in fructose ratio, the difference in viscosity, the difference in water content and the difference in hydroxymethylfurfural concentration.

[0101] It is understandable that the above-mentioned standard quality data is preset based on expert experience and stored in a database so as to be accessible in real time.

[0102] The method for obtaining a second comparison result based on the real-time surface image and the standard surface image includes:

[0103] The real-time surface image and the standard surface image are binarized respectively, and a first shadow area corresponding to the real-time surface image and a second shadow area corresponding to the standard surface image are calculated, and a ratio of the first shadow area to the second shadow area is taken as a second comparison result.

[0104] A weight generation module: used to determine a corresponding first weight based on a first comparison result, determine a corresponding second weight based on the first weight, determine a first vector according to the first comparison result and the first weight, and determine a second vector according to the second comparison result and the second weight;

[0105] The method for determining the corresponding first weight based on the first comparison result includes:

[0106]

[0107] In the formula, PRW is the first weight, Q rj is the initial value of the jth element in the first comparison result, Q sj is the standard value of the jth element in the first comparison result, n is the number of elements in the first comparison result, tan -1 (·) is the inverse tangent function, and π is a constant.

[0108] For example, if the first comparison result includes the difference in the fruit-glucose ratio, the difference in the viscosity value, the difference in the water content, and the difference in the hydroxymethylfurfural concentration, then n is 4, Q rj -Q sj is one of the above differences.

[0109] The method for determining a corresponding second weight based on a first weight includes:

[0110] SEW = 1-PRW;

[0111] Where SEW is the second weight.

[0112] The method for determining the first vector according to the first comparison result and the first weight includes:

[0113] The first comparison result is converted into a first comparison vector, the first weight is converted into a first weight vector, and the first comparison vector and the first weight vector are cascaded and fused to generate a first vector.

[0114] The method for determining the second vector according to the second comparison result and the second weight includes:

[0115] The second comparison result is converted into a second comparison vector, the second weight is converted into a second weight vector, and the second comparison vector and the second weight vector are cascaded and fused to generate a second vector.

[0116] Similarly, the implementation method is the same as the above content, and this embodiment will not go into details about it.

[0117] Improvement module: used for inputting the first vector and the second vector into the pre-built processing improvement model to obtain initial honey processing data.

[0118] In this embodiment, the initial honey processing data includes but is not limited to stirring time, heating temperature and pressurization time. It can be understood that the first comparison result includes the difference in fruit-to-glucose ratio, viscosity value difference, water content difference and hydroxymethylfurfural concentration difference.

[0119] The above-mentioned difference in the fructose ratio is related to the pressurization time. The fructose ratio refers to the ratio of the glucose content to the fructose content in honey. The crystallization of honey is also related to the glucose and fructose contents. At the same temperature, the solubility of glucose is much smaller than that of fructose, and the soluble glucose content is higher, and the honey crystallizes faster. Therefore, in this embodiment, the larger the difference in the fructose ratio, the higher the soluble glucose content of the unprocessed honey, and therefore the shorter the required pressurization time, because the pressurization time refers to the time for passing gas into the honey during the stirring process of the honey, and the purpose of pressurization is to make the honey crystallization effect better; it is understandable that those skilled in the art can set the standard fructose ratio to a lower value, then the initial fructose ratio is usually greater than the standard fructose ratio, and the fructose ratio difference is the initial fructose ratio minus the standard fructose ratio, so that the fructose ratio difference can characterize the soluble glucose content of the unprocessed honey.

[0120] The above-mentioned difference in viscosity values ​​is related to the stirring time. It can be understood that the larger the difference in viscosity values, the greater the viscosity of the unprocessed honey, and therefore the shorter the stirring time required.

[0121] The above-mentioned water content difference and hydroxymethylfurfural concentration difference are related to the heating temperature. The larger the water content difference, the higher the water content of the unprocessed honey, and the higher the required heating temperature; the larger the hydroxymethylfurfural concentration difference, the higher the hydroxymethylfurfural concentration of the unprocessed honey, and the lower the required heating temperature, because a higher heating temperature will increase the hydroxymethylfurfural concentration in the honey.

[0122] The second comparison result also affects the initial honey processing data. The closer the second comparison result is to 1, the better the quality of the honey, which means that the required stirring time and pressurizing time are shorter, and the required heating temperature is lower. As can be seen from the above content, the second comparison result is the ratio of the first shadow area to the second shadow area. The shadow area represents the impurities that are not filtered out of the honey, or the surface texture of the honey, etc. Similarly, if a standard surface image with a smaller second shadow area is pre-selected, the second comparison result will usually be greater than 1. The larger the second comparison result, the more unfiltered impurities in the surface honey or the more surface texture of the honey, which requires longer stirring time and pressurizing time, as well as higher heating temperature, because stirring and pressurizing are conducive to the floating of impurities, and higher heating temperature separates the fine particles in the honey from the impurities.

[0123] Correction module: used to obtain processing equipment data, correct the initial honey processing data according to the processing equipment data, and obtain target honey processing data. The processing equipment is the equipment required for processing honey.

[0124] It can be seen from the above content that the initial honey processing data includes but is not limited to stirring time, heating temperature and pressurizing time. In this embodiment, the processing equipment may be a high-pressure reactor, which can perform stirring, heating and pressurizing.

[0125] In this embodiment, the processing equipment data includes but is not limited to the number of maintenance records, the total operating time, the start-up time, and the start-up time. The start-up time is the sum of the stirring start-up time, the heating start-up time, and the pressurization start-up time. In this embodiment, the start-up time refers to the time required for the processing equipment to go from the starting working state to the stable working state. Taking the stirring start-up time as an example, the stirring start-up time refers to the time required from the start of rotation of the stirring shaft to the target speed. It can be understood that the above-mentioned start-up time is recorded by a sensor; still taking the stirring start-up time as an example, by setting a sensor on the stirring shaft, the start time is recorded from the beginning of rotation of the stirring shaft until the stirring shaft reaches the target speed, thereby ending the recording.

[0126] Methods for correcting initial honey processing data based on processing equipment data include:

[0127] The equipment aging coefficient is calculated based on the processing equipment data, and the initial honey processing data and the equipment aging coefficient are input into the aging correction model to obtain the target honey processing data.

[0128] Methods for calculating equipment aging coefficient based on processing equipment data include:

[0129]

[0130] EAC is the equipment aging coefficient, MRC is the number of maintenance records, TOE is the total operating time, STE is the start time, log 2 (·) is the logarithmic function with base 2, and cosh(·) represents the hyperbolic cosine function.

[0131] In this embodiment, the number of maintenance records, the total operating time, and the startup time are all positively correlated with the aging degree of the processing equipment. Therefore, it can be seen from the above formula that the equipment aging coefficient is positively correlated with the aging degree of the processing equipment.

[0132] Analysis module: used to obtain target quality data of processed honey, analyze according to the target quality data, and obtain analysis results. The processed honey is obtained by processing unprocessed honey according to the target honey processing data.

[0133] In this embodiment, the target quality data includes but is not limited to a target hydroxymethylfurfural concentration value, a target viscosity value, a target water content and a target conductivity. The target conductivity represents the mineral content in the honey. The larger the target conductivity, the more mineral content in the honey and the better the quality of the honey.

[0134] It should be noted that the analysis results include superior honey, first-grade honey and second-grade honey. Analysis based on target quality data means comparing the target quality data with the Chinese honey standard GB14963-2011, or with the international honey standard to obtain the analysis results.

[0135] Example 3

[0136] This embodiment discloses an electronic device, including a power supply, an interface, a keyboard, a memory, a central processing unit, and a computer program stored in the memory and executable on the central processing unit. When the central processing unit executes the computer program, a product quality analysis method based on industrial big data provided by the above methods is implemented. The interface includes a network interface and a data interface. The network interface includes a wired or wireless interface, and the data interface includes an input or output interface.

[0137] Since the electronic device introduced in this embodiment is an electronic device used to implement a product quality analysis method based on industrial big data in the embodiment of this application, based on the product quality analysis method based on industrial big data introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not described in detail here. As long as the technical personnel of this field implement the electronic device used in the product quality analysis method based on industrial big data in the embodiment of this application, it belongs to the scope of protection of this application.

[0138] Example 4

[0139] like Figure 3 As shown, this embodiment discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the above-mentioned product quality analysis method based on industrial big data is implemented.

[0140] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.

[0141] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or a wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD) or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0142] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0143] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0144] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0145] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0146] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0147] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0148] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A product quality analysis method based on industrial big data, characterized in that: include: S10: acquiring initial quality data and a real-time surface image of unprocessed honey, obtaining a first comparison result based on the initial quality data and the standard quality data, and obtaining a second comparison result based on the real-time surface image and the standard surface image; S20: determining a corresponding first weight based on the first comparison result, determining a corresponding second weight based on the first weight, determining a first vector based on the first comparison result and the first weight, and determining a second vector based on the second comparison result and the second weight; S30: inputting the first vector and the second vector into a pre-built processing improvement model to obtain initial honey processing data; S40: Acquire processing equipment data, and correct the initial honey processing data according to the processing equipment data to obtain target honey processing data, wherein the processing equipment is equipment required for processing honey; S50: Obtain target quality data of processed honey, perform analysis according to the target quality data, and obtain analysis results, wherein the processed honey is obtained by processing unprocessed honey according to the target honey processing data.

2. A product quality analysis method based on industrial big data according to claim 1, characterized in that: The method for obtaining a second comparison result based on the real-time surface image and the standard surface image comprises: The real-time surface image and the standard surface image are binarized respectively, and a first shadow area corresponding to the real-time surface image and a second shadow area corresponding to the standard surface image are calculated, and a ratio of the first shadow area to the second shadow area is taken as a second comparison result.

3. The product quality analysis method based on industrial big data according to claim 1 is characterized in that: The method for determining the corresponding first weight based on the first comparison result includes: In the formula, PRW is the first weight, Q rj is the initial value of the jth element in the first comparison result, Q sj is the standard value of the jth element in the first comparison result, n is the number of elements in the first comparison result, tan -1 (·) is the inverse tangent function, and π is a constant.

4. The product quality analysis method based on industrial big data according to claim 3 is characterized in that: The method for determining the first vector according to the first comparison result and the first weight includes: The first comparison result is converted into a first comparison vector, the first weight is converted into a first weight vector, and the first comparison vector and the first weight vector are cascaded and fused to generate a first vector.

5. The product quality analysis method based on industrial big data according to claim 1 is characterized in that: The method for constructing the processing improvement model includes: A processing sample data set is obtained, wherein the processing sample data set includes a first historical vector, a second historical vector and historical honey processing data, the processing sample data set is divided into a sample training set and a sample test set, a regression network is constructed, the first historical vector and the second historical vector in the sample training set are used as input data of the regression network, the historical honey processing data in the sample training set are used as output data of the regression network, the regression network is trained to obtain an initial regression network for predicting real-time honey processing data, the initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as a processing improvement model.

6. The product quality analysis method based on industrial big data according to claim 1 is characterized in that: The method for correcting the initial honey processing data according to the processing equipment data comprises: The equipment aging coefficient is calculated based on the processing equipment data, and the initial honey processing data and the equipment aging coefficient are input into the aging correction model to obtain the target honey processing data.

7. The product quality analysis method based on industrial big data according to claim 6 is characterized in that: The processing equipment data includes the number of maintenance records, the total operating time and the startup time. The method for calculating the equipment aging coefficient based on the processing equipment data includes: EAC is the equipment aging coefficient, MRC is the number of maintenance records, TOE is the total operating time, STE is the startup time, log2(·) is the logarithmic function with base 2, and cosh(·) represents the hyperbolic cosine function.

8. The product quality analysis method based on industrial big data according to claim 6 is characterized in that: The method for constructing the aging correction model comprises: An aging sample data set is obtained, wherein the aging sample data set includes historical initial honey processing data, historical equipment aging coefficients, and historical target honey processing data. The aging sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical initial honey processing data and the historical equipment aging coefficients in the sample training set are used as input data of the regression network, and the historical target honey processing data in the sample training set are used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting real-time target honey processing data. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as an aging correction model.

9. The product quality analysis method based on industrial big data according to claim 1 is characterized in that: The standard mass data includes a standard fruit-to-glucose ratio, a standard viscosity value, a standard water content, and a standard hydroxymethylfurfural concentration value. The method for obtaining the first comparison result based on the initial mass data and the standard mass data includes: The initial mass data and the standard mass data are traversed, the differences between the corresponding elements are calculated, a mass difference sequence is generated, and the mass difference sequence is used as the first comparison result. The mass difference sequence includes the difference in fructose ratio, the difference in viscosity, the difference in water content and the difference in hydroxymethylfurfural concentration.

10. A product quality analysis system based on industrial big data, which is used to implement a product quality analysis method based on industrial big data as claimed in any one of claims 1 to 9, characterized in that: include: A comparison generation module: used for obtaining initial quality data and a real-time surface image of unprocessed honey, obtaining a first comparison result based on the initial quality data and the standard quality data, and obtaining a second comparison result based on the real-time surface image and the standard surface image; A weight generation module: used to determine a corresponding first weight based on a first comparison result, determine a corresponding second weight based on the first weight, determine a first vector according to the first comparison result and the first weight, and determine a second vector according to the second comparison result and the second weight; Improvement module: used for inputting the first vector and the second vector into the pre-built processing improvement model to obtain initial honey processing data; Correction module: used to obtain processing equipment data, correct the initial honey processing data according to the processing equipment data, and obtain target honey processing data, wherein the processing equipment is the equipment required for processing honey; Analysis module: used to obtain target quality data of processed honey, analyze according to the target quality data, and obtain analysis results. The processed honey is obtained by processing unprocessed honey according to the target honey processing data.

Images