A method for calculating carbon footprint of product life cycle based on sustainable development

Abstract

The present invention relates to the technical field of carbon emission accounting, specifically a product life cycle carbon footprint accounting method based on sustainable development; collect and obtain a historical drying data set, the historical drying data set includes fabric input data, equipment drying parameters, drying carbon emission data and fabric output data; construct a drying effect recognition model to identify the fabric input image and the fabric drying image, and obtain a drying effect coefficient; set a drying effect threshold to screen the drying effect coefficient and historical drying data to obtain a drying data training set; obtain a unit carbon emission coefficient based on the drying carbon emission data and fabric input data in the drying data training set; obtain a drying parameter prediction model and a carbon emission accounting prediction model based on data training in the drying data training set. According to the drying parameter prediction model and the carbon emission accounting prediction model, the present invention can accurately obtain the equipment drying parameters with the best carbon emissions.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission accounting, and in particular to a method for calculating carbon footprint of a product life cycle based on sustainable development. Background Art

[0002] Fabric printing and dyeing is a process of applying color, pattern and other elements to fabrics. Its main purpose is to improve the appearance and performance of fabrics, make them more beautiful and personalized, and enhance the use value of fabrics. Fabric printing and dyeing is not only widely used in the field of clothing, but also occupies an important position in many industries such as home textiles and industrial fabrics.

[0003] The process of fabric printing and dyeing involves a large amount of energy consumption, chemical use and waste disposal, which will release a large amount of carbon emissions; and the drying stage is one of the key steps, which directly affects the energy consumption and carbon emissions of the entire process; the carbon emissions in the drying stage mainly come from the use of heat sources, which is used to quickly dry the fabric by heating, and this process consumes a lot of energy.

[0004] With the promotion of the concept of green development, more and more attention is paid to carbon emissions in the production process. By calculating the carbon footprint of the fabric printing and dyeing process, we can accurately understand the carbon emissions generated in each link and adjust the production process.

[0005] In the prior art, carbon emission accounting for the fabric drying process in the fabric printing and dyeing process is relatively rare, and most of the carbon emission accounting is carried out on the fabric drying process that has already occurred, and no predictive carbon emission accounting is carried out based on production parameters.

[0006] Therefore, a product life cycle carbon footprint accounting method based on sustainable development is proposed. Summary of the invention

[0007] The purpose of the present invention is to provide a product life cycle carbon footprint accounting method based on sustainable development, collect a historical drying data set, the historical drying data set includes fabric input data, equipment drying parameters, drying carbon emission data and fabric output data; construct a drying effect recognition model to identify the fabric input image and the fabric drying image to obtain a drying effect coefficient; set a drying effect threshold to screen the drying effect coefficient and the historical drying data to obtain a drying data training set; obtain a unit carbon emission coefficient based on the drying carbon emission data and fabric input data in the drying data training set; obtain a drying parameter prediction model and a carbon emission accounting prediction model based on the data training in the drying data training set. The present invention accurately obtains the equipment drying parameters with the best carbon emissions according to the drying parameter prediction model and the carbon emission accounting prediction model.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A product life cycle carbon footprint accounting method based on sustainable development, including:

[0010] S10. Collect historical drying data of the drying equipment to obtain a historical drying data set; the historical drying data set includes fabric input data, equipment drying parameters, drying carbon emission data and fabric output data; the fabric input data includes fabric input image, fabric material data, fabric weaving data, fabric shape parameters, printing pattern data and printing dye data; the fabric output data includes fabric drying image;

[0011] S20. Constructing a drying effect recognition model to recognize the cloth input image and the cloth drying image, and obtaining a drying effect coefficient according to the cloth drying quality, dye drying quality and pattern drying quality of the cloth drying image;

[0012] S30. Setting a drying effect threshold to filter the drying effect coefficient and historical drying data to obtain a drying data training set; calculating unit carbon emission data based on the drying carbon emission data and fabric input data in the drying data training set; calculating unit carbon emission coefficient based on the unit carbon emission data;

[0013] S40. According to the cloth input data, drying equipment parameters and unit carbon emission coefficient of the drying data training set, a drying parameter prediction model is trained; according to the cloth input data, drying equipment parameters and unit carbon emission data of the drying data training set, a carbon emission accounting prediction model is trained;

[0014] S50. Obtain input data of the fabric to be calculated and input it into a drying parameter prediction model to obtain drying equipment parameters to be calculated; identify the input data of the fabric to be calculated and the drying equipment parameters to be calculated through the carbon emission accounting prediction model to obtain carbon emission accounting data.

[0015] The drying equipment parameters include drying time data, drying humidity data, drying temperature data and drying wind speed data; the drying carbon emission data includes the carbon emissions generated by the fabric drying equipment during the drying process.

[0016] The drying effect recognition model includes a cloth drying recognition layer, a dye drying recognition layer, a pattern drying recognition layer and a drying effect recognition layer;

[0017] The cloth drying recognition layer uses the cloth input image as a reference to recognize the abnormal shape of the cloth in the cloth drying image, and calculates the cloth drying quality coefficient through the cloth flattening coefficient, the cloth defect coefficient and the cloth deformation coefficient;

[0018] The dye drying identification layer divides the cloth drying image into regions according to the color distribution to obtain color distribution sub-regions; using the cloth input image as a reference, the sub-region color coefficient is obtained through the color uniformity coefficient, color accuracy coefficient and color defect coefficient of the color distribution sub-region; and the dye drying quality coefficient is calculated according to the area proportion of the color distribution sub-region and the sub-region color coefficient;

[0019] The pattern drying recognition layer uses the cloth input image as a reference to recognize the pattern contour in the cloth drying image, and calculates the pattern drying quality coefficient according to the pattern integrity coefficient and the pattern deformation coefficient of the pattern contour;

[0020] The drying effect recognition layer obtains a drying effect coefficient according to the cloth drying quality coefficient, the dye drying quality coefficient and the pattern drying quality coefficient.

[0021] The calculation formula of the drying effect coefficient is:

[0022] ;

[0023] in, Indicates the drying effect coefficient; represents the first quality weight; Indicates the fabric drying quality coefficient; represents the second quality weight; Indicates the dye drying quality coefficient; represents the third quality weight; Indicates the pattern drying quality coefficient.

[0024] The calculation process of the unit carbon emission coefficient is as follows:

[0025] Obtain drying carbon emission data and fabric input data in the drying data training set;

[0026] The cloth area is calculated based on the cloth shape parameters in the cloth input data; the unit carbon emission data is calculated based on the drying carbon emission data and the cloth area;

[0027] All unit carbon emission data in the drying data training set are obtained, and the unit carbon emission coefficient is obtained through normalization.

[0028] The drying parameter prediction model includes a cloth data input layer, a cloth data recognition layer and a drying data prediction layer;

[0029] The fabric data input layer inputs the fabric input data to be calculated into the model, wherein the fabric input data to be calculated includes the fabric input image to be calculated, the fabric material data to be calculated, the fabric shape parameters to be calculated, the printing pattern data to be calculated and the printing dye data to be calculated;

[0030] The fabric data recognition layer extracts features of the fabric input data to be calculated to obtain feature data of the fabric to be calculated;

[0031] The drying data prediction layer identifies the feature data of the cloth to be calculated and obtains the parameters of the drying equipment to be calculated.

[0032] The carbon emission accounting prediction model includes a data input layer to be accounted for, a data identification layer to be accounted for, and a carbon emission data accounting layer;

[0033] The data input layer to be calculated inputs the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated into the model;

[0034] The data identification layer to be calculated performs feature extraction on the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated to obtain feature data of the cloth to be calculated and feature data of the equipment to be calculated;

[0035] The carbon emission data calculation layer identifies the characteristic data of the fabric to be calculated and the characteristic data of the equipment to be calculated to obtain the carbon emission calculation data.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention constructs a drying effect recognition model to recognize a cloth input image and a cloth drying image; the cloth drying quality coefficient is calculated through the cloth flattening coefficient, the cloth defect coefficient and the cloth deformation coefficient; the sub-region color coefficient is obtained through the color uniformity coefficient, the color accuracy coefficient and the color defect coefficient of the color distribution sub-region, and the dye drying quality coefficient is calculated according to the area proportion of the color distribution sub-region and the sub-region color coefficient; the pattern drying quality coefficient is calculated according to the pattern integrity coefficient and the pattern deformation coefficient of the pattern outline; the drying effect coefficient is accurately identified according to the cloth drying quality coefficient, the dye drying quality coefficient and the pattern drying quality coefficient.

[0038] 2. The present invention sets a drying effect threshold to screen the drying effect coefficient and historical drying data, and obtains a drying data training set based on the historical drying data set whose drying effect coefficient meets the conditions; based on the drying carbon emission data and fabric input data in the drying data training set, the unit carbon emission data and the unit carbon emission coefficient are calculated to provide an effective data basis for the training of the model.

[0039] 3. The present invention trains a drying parameter prediction model based on the cloth input data, drying equipment parameters and unit carbon emission coefficient of the drying data training set; identifies the cloth input data to be calculated through the drying parameter prediction model to obtain the cloth feature data to be calculated; identifies the cloth feature data to be calculated to obtain the drying equipment parameters to be calculated; and can accurately determine the drying equipment parameters that meet the requirements of the drying effect and have the optimal unit carbon emission coefficient based on the calculated cloth input data.

[0040] 4. The present invention trains a carbon emission accounting prediction model based on the cloth input data, drying equipment parameters and unit carbon emission data of the drying data training set; identifies the cloth input data to be calculated and the drying equipment parameters to be calculated through the carbon emission accounting prediction model to obtain the cloth feature data to be calculated and the equipment feature data to be calculated; and then accurately identifies the carbon emission accounting data through the cloth feature data to be calculated and the equipment feature data to be calculated. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a flow chart of a method for calculating carbon footprint of a product life cycle based on sustainable development according to the present invention;

[0042] Figure 2 It is a structural schematic diagram of the drying effect identification model of the present invention;

[0043] Figure 3 It is a structural schematic diagram of the drying parameter prediction model of the present invention;

[0044] Figure 4 It is a structural schematic diagram of the carbon emission accounting prediction model of the present invention. DETAILED DESCRIPTION

[0045] 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.

[0046] Embodiment 1

[0047] The present invention proposes a product life cycle carbon footprint accounting method based on sustainable development. The process of the method is as follows: Figure 1 As shown, including:

[0048] S10. Collect historical drying data of the drying equipment to obtain a historical drying data set; the historical drying data set includes fabric input data, equipment drying parameters, drying carbon emission data and fabric output data; the fabric input data includes fabric input image, fabric material data, fabric weaving data, fabric shape parameters, printing pattern data and printing dye data; the fabric output data includes fabric drying image.

[0049] Among them, the cloth input image is the image of the cloth before it is input into the drying equipment; the cloth material data is the material of the cloth, including cotton, polyester and wool, etc.; the cloth weaving data is the weaving method data of the cloth, including plain cloth, twill cloth, jacquard cloth and non-woven fabric; the cloth shape parameter is the shape of the cloth, such as the length and width of a rectangle; the radius of a circle, etc.; the printing pattern data is the pattern distribution of the cloth printing; the printing dye data is the dye data used in the pattern of the cloth printing.

[0050] The drying equipment parameters include drying time data, drying humidity data, drying temperature data and drying wind speed data; the drying carbon emission data includes the carbon emissions generated by the fabric drying equipment during the drying process.

[0051] The present invention collects historical drying data of the drying equipment to obtain a historical drying data set including fabric input data, equipment drying parameters, drying carbon emission data and fabric output data; through the historical drying data set, the historical working conditions of the drying equipment can be accurately and comprehensively understood, providing a basis for subsequent model construction and identification.

[0052] S20. Construct a drying effect recognition model to recognize the cloth input image and the cloth drying image, and obtain a drying effect coefficient according to the cloth drying quality, dye drying quality and pattern drying quality of the cloth drying image.

[0053] The structure of the drying effect identification model is as follows Figure 2 As shown; including a cloth drying identification layer, a dye drying identification layer, a pattern drying identification layer and a drying effect identification layer;

[0054] The cloth drying recognition layer uses the cloth input image as a reference to recognize the abnormal shape of the cloth in the cloth drying image, and calculates the cloth drying quality coefficient through the cloth flattening coefficient, the cloth defect coefficient and the cloth deformation coefficient; the cloth flattening coefficient is obtained through the area change of the cloth input image and the cloth drying image, and the texture recognition of the cloth drying image; the cloth defect coefficient is obtained through the feature recognition of the cloth drying image; the cloth deformation coefficient is obtained according to the shape change of the cloth input image and the cloth drying image;

[0055] The dye drying recognition layer divides the cloth drying image into regions according to the color distribution to obtain color distribution sub-regions; taking the cloth input image as a reference, the sub-region color coefficient is obtained through the color uniformity coefficient, color accuracy coefficient and color defect coefficient of the color distribution sub-region; the dye drying quality coefficient is calculated according to the area proportion of the color distribution sub-region and the sub-region color coefficient; the color uniformity coefficient is obtained by identifying the color distribution difference of the same color distribution sub-region of the cloth drying image; the color accuracy coefficient is obtained by identifying the color difference of the color sub-distribution sub-region corresponding to the cloth input image and the cloth drying image; the color defect coefficient is obtained by identifying the color texture feature of the color distribution sub-region;

[0056] The pattern drying recognition layer uses the cloth input image as a reference to recognize the pattern contour in the cloth drying image, and calculates the pattern drying quality coefficient according to the pattern integrity coefficient and the pattern deformation coefficient of the pattern contour; the pattern integrity coefficient is obtained according to the integrity of the pattern contour in the cloth drying image; the pattern deformation coefficient is obtained according to the difference in pattern contours between the cloth input image and the cloth drying image;

[0057] The drying effect recognition layer obtains a drying effect coefficient according to the cloth drying quality coefficient, the dye drying quality coefficient and the pattern drying quality coefficient.

[0058] The calculation formula of the drying effect coefficient is:

[0059] ;

[0060] in, Indicates the drying effect coefficient; represents the first quality weight; Indicates the fabric drying quality coefficient; represents the second quality weight; Indicates the dye drying quality coefficient; represents the third quality weight; Indicates the pattern drying quality coefficient.

[0061] The present invention constructs a drying effect recognition model to recognize a cloth input image and a cloth drying image; the cloth drying quality coefficient is calculated through the cloth flattening coefficient, the cloth defect coefficient and the cloth deformation coefficient; the sub-region color coefficient is obtained through the color uniformity coefficient, the color accuracy coefficient and the color defect coefficient of the color distribution sub-region, and the dye drying quality coefficient is calculated according to the area proportion of the color distribution sub-region and the sub-region color coefficient; the pattern drying quality coefficient is calculated according to the pattern integrity coefficient and the pattern deformation coefficient of the pattern outline; the drying effect coefficient is accurately recognized according to the cloth drying quality coefficient, the dye drying quality coefficient and the pattern drying quality coefficient.

[0062] S30. Set a drying effect threshold to filter the drying effect coefficient and historical drying data to obtain a drying data training set; calculate the unit carbon emission data based on the drying carbon emission data and fabric input data in the drying data training set; and calculate the unit carbon emission coefficient based on the unit carbon emission data.

[0063] The present invention sets a drying effect threshold to screen the drying effect coefficient and historical drying data, and obtains a drying data training set according to the historical drying data set whose drying effect coefficient meets the conditions; and obtains unit carbon emission data and unit carbon emission coefficient according to the drying carbon emission data and fabric input data in the drying data training set, thereby providing an effective data basis for model training.

[0064] The calculation process of the unit carbon emission coefficient is as follows:

[0065] Obtain drying carbon emission data and fabric input data in the drying data training set;

[0066] The cloth area is calculated based on the cloth shape parameters in the cloth input data; the unit carbon emission data is calculated based on the drying carbon emission data and the cloth area;

[0067] All unit carbon emission data in the drying data training set are obtained, and the unit carbon emission coefficient is obtained through normalization.

[0068] The present invention obtains drying carbon emission data and cloth input data in a drying data training set; calculates the cloth area according to the cloth shape parameters in the cloth input data; calculates the unit carbon emission data according to the drying carbon emission data and the cloth area; performs normalization processing on the unit carbon emission data to obtain the unit carbon emission coefficient; through the unit carbon emission coefficient, the carbon emission situation of the cloth input data and the equipment drying parameters can be accurately identified and measured.

[0069] S40. According to the cloth input data, drying equipment parameters and unit carbon emission coefficient of the drying data training set, a drying parameter prediction model is trained; according to the cloth input data, drying equipment parameters and unit carbon emission data of the drying data training set, a carbon emission accounting prediction model is trained.

[0070] The structure of the drying parameter prediction model is as follows Figure 3 As shown; including a cloth data input layer, a cloth data recognition layer and a drying data prediction layer;

[0071] The fabric data input layer inputs the fabric input data to be calculated into the model, wherein the fabric input data to be calculated includes the fabric input image to be calculated, the fabric material data to be calculated, the fabric shape parameters to be calculated, the printing pattern data to be calculated and the printing dye data to be calculated;

[0072] The fabric data recognition layer extracts features of the input data of the fabric to be calculated to obtain feature data of the fabric to be calculated;

[0073] The drying data prediction layer identifies the feature data of the cloth to be calculated and obtains the parameters of the drying equipment to be calculated.

[0074] The present invention trains a drying parameter prediction model according to the cloth input data, drying equipment parameters and unit carbon emission coefficient of a drying data training set; identifies the cloth input data to be calculated by the drying parameter prediction model to obtain the cloth feature data to be calculated; identifies the cloth feature data to be calculated to obtain the drying equipment parameters to be calculated; and can accurately determine the drying equipment parameters with the drying effect meeting the requirements and the lowest carbon emission coefficient according to the calculated cloth input data.

[0075] The structure of the carbon emission accounting prediction model is as follows Figure 4 As shown; including the data input layer to be accounted for, the data identification layer to be accounted for and the carbon emission data accounting layer;

[0076] The data input layer to be calculated inputs the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated into the model;

[0077] The data identification layer to be calculated performs feature extraction on the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated to obtain feature data of the cloth to be calculated and feature data of the equipment to be calculated;

[0078] The carbon emission data calculation layer identifies the characteristic data of the fabric to be calculated and the characteristic data of the equipment to be calculated to obtain the carbon emission calculation data.

[0079] The present invention trains a carbon emission accounting prediction model based on the cloth input data, drying equipment parameters and unit carbon emission data of a drying data training set; identifies the cloth input data to be calculated and the drying equipment parameters to be calculated through the carbon emission accounting prediction model to obtain the cloth feature data to be calculated and the equipment feature data to be calculated; and then accurately identifies the carbon emission accounting data through the cloth feature data to be calculated and the equipment feature data to be calculated.

[0080] S50. Obtain input data of the fabric to be calculated and input it into a drying parameter prediction model to obtain drying equipment parameters to be calculated; identify the input data of the fabric to be calculated and the drying equipment parameters to be calculated through the carbon emission accounting prediction model to obtain carbon emission accounting data.

[0081] The present invention collects a historical drying data set, which includes cloth input data, equipment drying parameters, drying carbon emission data and cloth output data; constructs a drying effect recognition model to recognize cloth input images and cloth drying images, and obtains a drying effect coefficient; sets a drying effect threshold to screen the drying effect coefficient and historical drying data, and obtains a drying data training set; obtains a unit carbon emission coefficient according to the drying carbon emission data and cloth input data in the drying data training set; obtains a drying parameter prediction model and a carbon emission accounting prediction model according to data training in the drying data training set, and accurately obtains the equipment drying parameters with the optimal carbon emissions.

[0082] Embodiment 2

[0083] With the promotion of the concept of green development, more and more attention has been paid to the carbon emissions in the process of fabric printing and dyeing. By calculating the carbon footprint, we can accurately understand the carbon emissions generated in each link and adjust the production process. To this end, a product life cycle carbon footprint calculation method based on sustainable development is proposed. The process of the method is as follows: Figure 1 As shown, including:

[0084] S10. Collect historical drying data of the drying equipment to obtain a historical drying data set; the historical drying data set includes fabric input data, equipment drying parameters, drying carbon emission data and fabric output data; the fabric input data includes fabric input image, fabric material data, fabric weaving data, fabric shape parameters, printing pattern data and printing dye data; the fabric output data includes fabric drying image.

[0085] The drying equipment parameters include drying time data, drying humidity data, drying temperature data and drying wind speed data; the drying carbon emission data includes the carbon emissions generated by the fabric drying equipment during the drying process.

[0086] Select the cloth input data at a certain moment, as shown in Table 1.

[0087] Table 1 Fabric input data table

[0088]

[0089] The present invention collects historical drying data of the drying equipment to obtain a historical drying data set including fabric input data, equipment drying parameters, drying carbon emission data and fabric output data; through the historical drying data set, the historical working conditions of the drying equipment can be accurately and comprehensively understood, providing a basis for subsequent model construction and identification.

[0090] S20. Construct a drying effect recognition model to recognize the cloth input image and the cloth drying image, and obtain a drying effect coefficient according to the cloth drying quality, dye drying quality and pattern drying quality of the cloth drying image.

[0091] The structure of the drying effect identification model is as follows Figure 2 As shown; including a cloth drying identification layer, a dye drying identification layer, a pattern drying identification layer and a drying effect identification layer;

[0092] The cloth drying recognition layer uses the cloth input image as a reference to recognize the abnormal shape of the cloth in the cloth drying image, and calculates the cloth drying quality coefficient through the cloth flattening coefficient, the cloth defect coefficient and the cloth deformation coefficient; the cloth flattening coefficient is obtained through the area change of the cloth input image and the cloth drying image, and the texture recognition of the cloth drying image; the cloth defect coefficient is obtained through the feature recognition of the cloth drying image; the cloth deformation coefficient is obtained according to the shape change of the cloth input image and the cloth drying image;

[0093] The dye drying recognition layer divides the cloth drying image into regions according to the color distribution to obtain color distribution sub-regions; taking the cloth input image as a reference, the sub-region color coefficient is obtained through the color uniformity coefficient, color accuracy coefficient and color defect coefficient of the color distribution sub-region; the dye drying quality coefficient is calculated according to the area proportion of the color distribution sub-region and the sub-region color coefficient; the color uniformity coefficient is obtained by identifying the color distribution difference of the same color distribution sub-region of the cloth drying image; the color accuracy coefficient is obtained by identifying the color difference of the color sub-distribution sub-region corresponding to the cloth input image and the cloth drying image; the color defect coefficient is obtained by identifying the color texture feature of the color distribution sub-region;

[0094] The pattern drying recognition layer uses the cloth input image as a reference to recognize the pattern contour in the cloth drying image, and calculates the pattern drying quality coefficient according to the pattern integrity coefficient and the pattern deformation coefficient of the pattern contour; the pattern contour integrity coefficient is obtained according to the integrity of the pattern contour in the cloth drying image; the pattern deformation coefficient is obtained according to the difference in pattern contours between the cloth input image and the cloth drying image;

[0095] The drying effect recognition layer obtains a drying effect coefficient according to the fabric drying quality coefficient, the dye drying quality coefficient and the pattern drying quality coefficient. The calculation formula of the drying effect coefficient is:

[0096] ;

[0097] in, Indicates the drying effect coefficient; represents the first quality weight; Indicates the fabric drying quality coefficient; represents the second quality weight; Indicates the dye drying quality coefficient; represents the third quality weight; Indicates the pattern drying quality coefficient.

[0098] In order to verify the recognition effect of the drying effect recognition model of the present invention, it is verified;

[0099] Four verification models are constructed during the verification process, including verification model 1, verification model 2, verification model 3 and verification model 4; the verification model 1 is the drying effect recognition model of the present invention, which obtains the first drying effect coefficient through three dimensions of cloth drying recognition, dye drying recognition and pattern drying recognition;

[0100] The verification model 2, based on the verification model 1, does not consider the fabric drying recognition, and only obtains the second drying effect coefficient through the dye drying recognition and the pattern drying recognition;

[0101] The verification model 3, based on the verification model 1, does not consider the dye drying recognition, and only obtains the third drying effect coefficient through the cloth drying recognition and the pattern drying recognition;

[0102] The verification model 4, based on the verification model 1, does not consider pattern drying recognition, and only obtains the fourth drying effect coefficient through cloth drying recognition and dye drying recognition.

[0103] The recognition accuracy of the model is obtained according to the degree of proximity between the first drying effect coefficient, the second drying effect coefficient, the third drying effect coefficient and the fourth drying effect coefficient and the effect coefficient label respectively.

[0104] The obtained data are shown in Table 2.

[0105] Table 2 Validation data table of drying effect identification model

[0106]

[0107] From the data in Table 2, it can be seen that the recognition accuracy of verification model 1 is the highest.

[0108] The present invention constructs a drying effect recognition model to recognize a cloth input image and a cloth drying image; the cloth drying quality coefficient is calculated through the cloth flattening coefficient, the cloth defect coefficient and the cloth deformation coefficient; the sub-region color coefficient is obtained through the color uniformity coefficient, the color accuracy coefficient and the color defect coefficient of the color distribution sub-region, and the dye drying quality coefficient is calculated according to the area proportion of the color distribution sub-region and the sub-region color coefficient; the pattern drying quality coefficient is calculated according to the pattern integrity coefficient and the pattern deformation coefficient of the pattern outline; the drying effect coefficient is accurately recognized according to the cloth drying quality coefficient, the dye drying quality coefficient and the pattern drying quality coefficient.

[0109] S30. Set a drying effect threshold to filter the drying effect coefficient and historical drying data to obtain a drying data training set; calculate the unit carbon emission data based on the drying carbon emission data and fabric input data in the drying data training set; and calculate the unit carbon emission coefficient based on the unit carbon emission data.

[0110] The present invention sets a drying effect threshold to screen the drying effect coefficient and historical drying data, and obtains a drying data training set according to the historical drying data set whose drying effect coefficient meets the conditions; and obtains unit carbon emission data and unit carbon emission coefficient according to the drying carbon emission data and fabric input data in the drying data training set, thereby providing an effective data basis for model training.

[0111] The calculation process of the unit carbon emission coefficient is as follows:

[0112] Obtain drying carbon emission data and fabric input data in the drying data training set;

[0113] The cloth area is calculated based on the cloth shape parameters in the cloth input data; the unit carbon emission data is calculated based on the drying carbon emission data and the cloth area;

[0114] All unit carbon emission data in the drying data training set are obtained, and the unit carbon emission coefficient is obtained through normalization.

[0115] The present invention obtains drying carbon emission data and cloth input data in a drying data training set; calculates the cloth area according to the cloth shape parameters in the cloth input data; calculates the unit carbon emission data according to the drying carbon emission data and the cloth area; performs normalization processing on the unit carbon emission data to obtain the unit carbon emission coefficient; through the unit carbon emission coefficient, the carbon emission situation of the cloth input data and the equipment drying parameters can be accurately identified and measured.

[0116] S40. According to the cloth input data, drying equipment parameters and unit carbon emission coefficient of the drying data training set, a drying parameter prediction model is trained; according to the cloth input data, drying equipment parameters and unit carbon emission data of the drying data training set, a carbon emission accounting prediction model is trained.

[0117] The structure of the drying parameter prediction model is as follows Figure 3 As shown; including a cloth data input layer, a cloth data recognition layer and a drying data prediction layer;

[0118] The fabric data input layer inputs the fabric input data to be calculated into the model, wherein the fabric input data to be calculated includes the fabric input image to be calculated, the fabric material data to be calculated, the fabric shape parameters to be calculated, the printing pattern data to be calculated and the printing dye data to be calculated;

[0119] The fabric data recognition layer extracts features of the input data of the fabric to be calculated to obtain feature data of the fabric to be calculated;

[0120] The drying data prediction layer identifies the feature data of the cloth to be calculated and obtains the parameters of the drying equipment to be calculated.

[0121] The present invention trains a drying parameter prediction model according to the cloth input data, drying equipment parameters and unit carbon emission coefficient of a drying data training set; identifies the cloth input data to be calculated by the drying parameter prediction model to obtain the cloth feature data to be calculated; identifies the cloth feature data to be calculated to obtain the drying equipment parameters to be calculated; and can accurately determine the drying equipment parameters with the drying effect meeting the requirements and the lowest carbon emission coefficient according to the calculated cloth input data.

[0122] The structure of the carbon emission accounting prediction model is as follows Figure 4 As shown; including the data input layer to be accounted for, the data identification layer to be accounted for and the carbon emission data accounting layer;

[0123] The data input layer to be calculated inputs the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated into the model;

[0124] The data identification layer to be calculated performs feature extraction on the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated to obtain feature data of the cloth to be calculated and feature data of the equipment to be calculated;

[0125] The carbon emission data calculation layer identifies the characteristic data of the fabric to be calculated and the characteristic data of the equipment to be calculated to obtain the carbon emission calculation data.

[0126] The present invention trains a carbon emission accounting prediction model based on the cloth input data, drying equipment parameters and unit carbon emission data of a drying data training set; identifies the cloth input data to be calculated and the drying equipment parameters to be calculated through the carbon emission accounting prediction model to obtain the cloth feature data to be calculated and the equipment feature data to be calculated; and then accurately identifies the carbon emission accounting data through the cloth feature data to be calculated and the equipment feature data to be calculated.

[0127] S50. Obtain input data of the fabric to be calculated and input it into a drying parameter prediction model to obtain drying equipment parameters to be calculated; identify the input data of the fabric to be calculated and the drying equipment parameters to be calculated through the carbon emission accounting prediction model to obtain carbon emission accounting data.

[0128] The present invention collects a historical drying data set, which includes cloth input data, equipment drying parameters, drying carbon emission data and cloth output data; constructs a drying effect recognition model to recognize cloth input images and cloth drying images, and obtains a drying effect coefficient; sets a drying effect threshold to screen the drying effect coefficient and historical drying data, and obtains a drying data training set; obtains a unit carbon emission coefficient according to the drying carbon emission data and cloth input data in the drying data training set; obtains a drying parameter prediction model and a carbon emission accounting prediction model according to data training in the drying data training set, and accurately obtains the equipment drying parameters with the optimal carbon emissions.

[0129] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A product life cycle carbon footprint accounting method based on sustainable development, characterized in that: include: S10. Collect historical drying data sets of drying equipment, including cloth input data, equipment drying parameters, drying carbon emission data and cloth output data; The cloth input data includes cloth input image, cloth material data, cloth weaving data, cloth shape parameters, printing pattern data and printing dye data; the cloth output data includes cloth drying image; S20. Constructing a drying effect recognition model to recognize the cloth input image and the cloth drying image, and obtaining a drying effect coefficient according to the cloth drying quality, dye drying quality and pattern drying quality of the cloth drying image; S30. Setting a drying effect threshold to filter the drying effect coefficient and historical drying data to obtain a drying data training set; calculating unit carbon emission data based on the drying carbon emission data and fabric input data in the drying data training set; The unit carbon emission coefficient is calculated based on the unit carbon emission data; S40. According to the cloth input data, drying equipment parameters and unit carbon emission coefficient of the drying data training set, a drying parameter prediction model is trained; according to the cloth input data, drying equipment parameters and unit carbon emission data of the drying data training set, a carbon emission accounting prediction model is trained; S50. Obtain input data of the fabric to be calculated and input it into a drying parameter prediction model to obtain drying equipment parameters to be calculated; identify the input data of the fabric to be calculated and the drying equipment parameters to be calculated through the carbon emission accounting prediction model to obtain carbon emission accounting data.

2. The method for calculating carbon footprint of a product life cycle based on sustainable development according to claim 1, characterized in that: The drying equipment parameters include drying time data, drying humidity data, drying temperature data and drying wind speed data; the drying carbon emission data includes the carbon emissions generated by the fabric drying equipment during the drying process.

3. The method for calculating carbon footprint of a product life cycle based on sustainable development according to claim 1, characterized in that: The drying effect recognition model includes a cloth drying recognition layer, a dye drying recognition layer, a pattern drying recognition layer and a drying effect recognition layer; The cloth drying recognition layer uses the cloth input image as a reference to recognize the abnormal shape of the cloth in the cloth drying image, and calculates the cloth drying quality coefficient through the cloth flattening coefficient, the cloth defect coefficient and the cloth deformation coefficient; The dye drying identification layer divides the cloth drying image into regions according to the color distribution to obtain color distribution sub-regions; taking the cloth input image as a reference, the sub-region color coefficient is obtained through the color uniformity coefficient, color accuracy coefficient and color defect coefficient of the color distribution sub-region; The dye drying quality coefficient is calculated based on the area proportion of the color distribution sub-region and the sub-region color coefficient; The pattern drying recognition layer uses the cloth input image as a reference to recognize the pattern contour in the cloth drying image, and calculates the pattern drying quality coefficient according to the pattern integrity coefficient and the pattern deformation coefficient of the pattern contour; The drying effect recognition layer obtains a drying effect coefficient according to the cloth drying quality coefficient, the dye drying quality coefficient and the pattern drying quality coefficient.

4. The method for calculating carbon footprint of a product life cycle based on sustainable development according to claim 3 is characterized by: The calculation formula of the drying effect coefficient is: ； in, Indicates the drying effect coefficient; represents the first quality weight; Indicates the fabric drying quality coefficient; represents the second quality weight; Indicates the dye drying quality coefficient; represents the third quality weight; Indicates the pattern drying quality coefficient.

5. The method for calculating carbon footprint of a product life cycle based on sustainable development according to claim 1, characterized in that: The calculation process of the unit carbon emission coefficient is as follows: Obtain drying carbon emission data and fabric input data in the drying data training set; The cloth area is calculated based on the cloth shape parameters in the cloth input data; the unit carbon emission data is calculated based on the drying carbon emission data and the cloth area; All unit carbon emission data in the drying data training set are obtained, and the unit carbon emission coefficient is obtained through normalization.

6. The method for calculating carbon footprint of a product life cycle based on sustainable development according to claim 1, characterized in that: The drying parameter prediction model includes a cloth data input layer, a cloth data recognition layer and a drying data prediction layer; The fabric data input layer inputs the fabric input data to be calculated into the model, wherein the fabric input data to be calculated includes the fabric input image to be calculated, the fabric material data to be calculated, the fabric shape parameters to be calculated, the printing pattern data to be calculated and the printing dye data to be calculated; The fabric data recognition layer extracts features of the input data of the fabric to be calculated to obtain feature data of the fabric to be calculated; The drying data prediction layer identifies the feature data of the cloth to be calculated and obtains the parameters of the drying equipment to be calculated.

7. The method for calculating carbon footprint of a product life cycle based on sustainable development according to claim 1 is characterized by: The carbon emission accounting prediction model includes a data input layer to be accounted for, a data identification layer to be accounted for, and a carbon emission data accounting layer; The data input layer to be calculated inputs the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated into the model; The data identification layer to be calculated performs feature extraction on the input data of the cloth to be calculated and the parameters of the drying equipment to be calculated to obtain feature data of the cloth to be calculated and feature data of the equipment to be calculated; The carbon emission data calculation layer identifies the characteristic data of the fabric to be calculated and the characteristic data of the equipment to be calculated to obtain the carbon emission calculation data.

Images