Cost control method and system for industrial solid waste treatment

Abstract

The present application relates to a kind of industrial solid waste processing cost control method and system, comprising the following steps, by multispectral imaging equipment to industrial solid waste is scanned, obtains solid waste multispectral image data;The band response characteristic extraction of the solid waste multispectral image data is obtained Material spectral fingerprint feature;Based on the material spectral fingerprint feature, the composition of the industrial solid waste is determined, and the solid waste composition vector is obtained;Based on the solid waste composition vector, the process correlation of the industrial solid waste is obtained Component-process matching weight matrix;Based on the component-process matching weight matrix, global cost control is carried out, and the target cost processing strategy scheme is obtained, solves the technical problem that the complex correlation between different components and processing technology is often ignored in the prior art, so that the economic efficiency and feasibility of different processing schemes cannot be accurately evaluated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial solid waste, in particular to a cost control method and system for industrial solid waste treatment. BACKGROUND

[0002] With the rapid development of industrialization and urbanization, the amount of industrial solid waste is increasing year by year, and its treatment and resource utilization has become an important issue for environmental protection and sustainable development. Although some studies have attempted to introduce spectral analysis technology for component identification of solid waste, most of them are still in the laboratory stage and have not been able to achieve large-scale industrial application. In particular, there is still a lack of a complete theoretical framework and technical path on how to effectively match the spectral identification results with subsequent treatment processes. Existing methods often ignore the complex correlation between different components and treatment processes, resulting in the inability to accurately assess the economic feasibility and feasibility of different treatment schemes, thereby affecting the overall cost control effect. SUMMARY

[0003] The main purpose of the present application is to provide a cost control method for industrial solid waste treatment, which solves the technical problem that existing methods often ignore the complex correlation between different components and treatment processes, resulting in the inability to accurately assess the economic feasibility and feasibility of different treatment schemes.

[0004] To achieve the above purpose, the present application provides a cost control method for industrial solid waste treatment, comprising the following steps:

[0005] Performing multi-spectral scanning on the industrial solid waste by a multi-spectral imaging device to obtain solid waste multi-spectral image data;

[0006] Performing waveband response feature extraction on the solid waste multi-spectral image data to obtain material spectral fingerprint features;

[0007] Determining the components of the industrial solid waste based on the material spectral fingerprint features to obtain a solid waste component vector;

[0008] Correlating the industrial solid waste based on the solid waste component vector to obtain a component-process matching weight matrix;

[0009] Performing global cost control based on the component-process matching weight matrix to obtain a target cost treatment strategy.

[0010] Further, the multi-spectral scanning on the industrial solid waste by the multi-spectral imaging device to obtain the solid waste multi-spectral image data comprises:

[0011] Scanning the surface and internal structure of the industrial solid waste by the multi-spectral imaging device to obtain multi-band reflection and transmission image data;

[0012] correcting spatial resolution in the multi-band reflectance and transmittance image data to generate corrected multi-spectral image data;

[0013] performing spectral feature enhancement on the corrected multi-spectral image data to obtain solid waste multi-spectral image data.

[0014] Further, the waveband response feature extraction on the solid waste multi-spectral image data to obtain material spectral fingerprint features includes:

[0015] performing inter-waveband correlation filtering on the solid waste multi-spectral image data, and performing spatial domain frequency decomposition on the filtered solid waste multi-spectral image data to obtain spectral response data;

[0016] extracting local waveband features in the spectral response data by an adaptive window sliding method to obtain a waveband local extreme point sequence, and performing spectral absorption peak position identification based on the waveband local extreme point sequence to obtain feature peak position coordinates;

[0017] dividing waveband intervals of the spectral response data based on the feature peak position coordinates to obtain a plurality of feature waveband sub-intervals, and calculating spectral reflectance gradients of the feature waveband sub-intervals to obtain a waveband gradient feature vector;

[0018] performing spectral feature dimension reduction based on the waveband gradient feature vector to obtain a low-dimensional spectral feature matrix, and performing eigenvalue decomposition on the low-dimensional spectral feature matrix to obtain material spectral fingerprint features.

[0019] Further, the component determination of the industrial solid waste based on the material spectral fingerprint features to obtain a solid waste component vector includes:

[0020] performing spectral line contrast calibration on the material spectral fingerprint features to obtain chemical element characteristic spectral data, and analyzing material components in the chemical element characteristic spectral data to obtain an element composition distribution matrix;

[0021] performing element quantitative analysis on the element composition distribution matrix to obtain an element content distribution atlas, and performing spatial clustering analysis on the element content distribution atlas to obtain element enrichment area parameters;

[0022] deducing compound structures of the industrial solid waste based on the element enrichment area parameters to generate a material form combination list, and performing stoichiometric balance verification on the material form combination list to obtain a component proportion correction coefficient;

[0023] determining components of the industrial solid waste based on the component proportion correction coefficient by a material balance constraint method to obtain a solid waste component vector.

[0024] Further, the element composition distribution matrix is subjected to element quantitative analysis to obtain an element content distribution spectrum, comprising:

[0025] The element composition distribution matrix is subjected to atomic emission spectrum intensity correction to obtain an element characteristic spectrum line intensity table, and the element characteristic spectrum line intensity table is subjected to matrix effect compensation calculation to obtain an element sensitivity coefficient matrix;

[0026] The element sensitivity coefficient matrix is subjected to interference peak unfolding by an element internal standard method to obtain an element pure spectrum, and the element pure spectrum is subjected to spectrum peak area integration to generate an element content calibration curve;

[0027] Based on the element content calibration curve, regional element concentration mapping is performed to obtain an element spatial distribution vector group, and the element spatial distribution vector group is subjected to element correlation clustering to generate an element paragenetic relationship spectrum;

[0028] Based on the element paragenetic relationship spectrum, element partition coefficient calculation is performed to obtain an element occurrence state mapping diagram, and the element occurrence state mapping diagram is subjected to chemical valence state balance constraint to obtain an element content distribution spectrum.

[0029] Further, the industrial solid waste is subjected to process correlation based on the solid waste composition vector to obtain a composition-process matching weight matrix, comprising:

[0030] The solid waste composition vector is subjected to physicochemical property analysis to obtain a solid waste reaction activity parameter group, and the industrial solid waste is subjected to thermodynamic stability analysis based on the solid waste reaction activity parameter group to obtain a solid waste thermodynamic stability matrix;

[0031] The industrial solid waste is subjected to conversion rate prediction by the solid waste thermodynamic stability matrix to obtain a conversion rate vector, and a composition-process matching weight matrix is obtained based on the conversion rate vector.

[0032] Further, the global cost calculation control is performed based on the composition-process matching weight matrix to obtain a target cost treatment strategy scheme, comprising:

[0033] The composition-process matching weight matrix is subjected to dynamic programming decomposition to obtain a process unit cost constraint condition set, and process scheduling optimization is performed based on the process unit cost constraint condition set to obtain a treatment process timing arrangement table;

[0034] The treatment process timing arrangement table is subjected to resource configuration calculation by multi-stage cost accumulation to obtain a resource scheduling strategy matrix, and the resource scheduling strategy matrix is subjected to boundary constraint verification to obtain a process operation parameter configuration set;

[0035] The cost control index decomposition is performed based on the process operation parameter configuration set, and a target cost treatment strategy scheme is obtained.

[0036] The application further provides an industrial solid waste treatment cost control system, comprising:

[0037] The scanning module is configured to perform multispectral scanning on the industrial solid waste by using the multispectral imaging device to obtain multispectral image data of the solid waste.

[0038] The extraction module is configured to extract a substance spectral fingerprint feature from the multispectral image data of the solid waste.

[0039] The determination module is configured to determine the composition of the industrial solid waste based on the substance spectral fingerprint feature to obtain a solid waste composition vector.

[0040] The correlation module is configured to correlate the industrial solid waste with a process based on the solid waste composition vector to obtain a composition-process matching weight matrix.

[0041] The control module is configured to perform global cost control based on the composition-process matching weight matrix to obtain a target cost treatment strategy scheme.

[0042] The application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method according to any one of the preceding embodiments when executing the computer program.

[0043] The application further provides a computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the steps of the method according to any one of the preceding embodiments.

[0044] The application provides an industrial solid waste treatment cost control method, which comprises the following steps: performing multispectral scanning on industrial solid waste by using a multispectral imaging device to obtain multispectral image data of the solid waste; extracting a substance spectral fingerprint feature from the multispectral image data of the solid waste; determining the composition of the industrial solid waste based on the substance spectral fingerprint feature to obtain a solid waste composition vector; correlating the industrial solid waste with a process based on the solid waste composition vector to obtain a composition-process matching weight matrix; and performing global cost control based on the composition-process matching weight matrix to obtain a target cost treatment strategy scheme. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 is a schematic diagram of the steps of the cost control method for industrial solid waste treatment in an embodiment of the present application;

[0046] Figure 2 is a structural block diagram of the cost control system for industrial solid waste treatment in an embodiment of the present application;

[0047] Figure 3 is a structural schematic block diagram of a computer device of an embodiment of the present application.

[0048] The purposes, functional features and advantages of the present application will be further described with reference to the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0049] In order to make the purposes, technical solutions and advantages of the present application clearer, further detailed description will be made of the present application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0050] As shown in Figure 1 , the present application is a cost control method for industrial solid waste treatment, comprising the following steps: Figure 1

[0051] Step S1, multi-spectral scanning of industrial solid waste is performed by a multi-spectral imaging device to obtain solid waste multi-spectral image data.

[0052] Specifically, multi-spectral scanning of industrial solid waste is performed by a multi-spectral imaging device to obtain solid waste multi-spectral image data, which is a basic step for realizing component identification and subsequent process matching in this method. In actual operation, industrial solid waste usually exists in the form of a complex mixture, and its physical form and chemical composition have high heterogeneity. Traditional single waveband imaging or manual sampling analysis cannot fully and accurately reflect its internal composition. Therefore, the multi-spectral imaging device is used to perform non-contact multi-spectral scanning of the solid waste, which can simultaneously obtain its surface morphology information and reflection characteristic data under multiple wavebands, thereby forming a multi-dimensional image data set containing rich spectral information. For example, in the solid waste treatment scene of the steel and iron metallurgical industry, the device can perform online scanning on the crushed waste slag, capture the response differences of different mineral components at multiple specific wavelengths, and then construct a high-resolution solid waste multi-spectral image data, providing a high-quality data basis for subsequent feature extraction and component identification. This process not only improves the detection efficiency, but also significantly enhances the accuracy and stability of complex component identification.

[0053] Step S2, waveband response feature extraction is performed on the solid waste multi-spectral image data to obtain material spectral fingerprint features.

[0054] ​Specifically, the band response feature extraction of the solid waste multi-spectral image data to obtain the material spectral fingerprint feature is a key technical link to realize the identification of industrial solid waste composition. After obtaining the solid waste multi-spectral image data collected by the multi-spectral imaging device, further digital image processing and spectral analysis are required to extract the reflection or absorption response rules of different materials to specific electromagnetic waves under each wave band. This process usually includes image preprocessing, wave band selection optimization, spectral feature extraction and other steps. Through algorithm model analysis on the multi-wave band response value of each pixel point, the spectral fingerprint feature vector reflecting the essential properties of the material is constructed. For example, in the process of solid waste treatment generated in the steel metallurgical industry, the image data obtained by multi-spectral scanning contains information of iron oxides, silicates, carbides and other mineral components. The reflectivity of these components under different wave bands has slight differences. Through modeling and feature extraction of these differences, the spectral fingerprint features of various materials can be accurately identified, laying a foundation for subsequent quantitative and qualitative analysis of composition. This process not only improves the recognition accuracy, but also provides reliable data support for subsequent process matching.

[0055] Step S3, determining the composition of the industrial solid waste based on the material spectral fingerprint feature to obtain a solid waste composition vector.

[0056] Specifically, determining the composition of the industrial solid waste based on the material spectral fingerprint feature to obtain a solid waste composition vector is a key link to convert spectral information into quantifiable and analyzable composition data. After completing the extraction of the spectral fingerprint features of each region in the solid waste multi-spectral image data, the system will compare and match the extracted spectral features with the standard spectra of known materials through the pre-constructed spectral database or machine learning classification model, to identify the specific types of materials contained in the industrial solid waste, and further estimate the relative content of each component by combining the relationship model between spectral response intensity and material content, to finally form a solid waste composition vector represented in numerical form. For example, in the process of solid waste treatment generated in the steel metallurgical industry, the system can identify the main components such as iron oxides, silicates and calcium carbonate contained in the waste slag according to the extracted spectral fingerprint features, and obtain the proportion of each component in the whole through quantitative analysis, to further generate a composition vector containing the content information of multiple materials, providing accurate data support for subsequent process selection. This step not only realizes the fine identification of complex solid waste composition, but also lays a solid foundation for establishing the correlation between composition and processing technology.

[0057] Step S4, process correlation of the industrial solid waste based on the solid waste composition vector to obtain a composition-process matching weight matrix.

[0058] Specifically, based on the solid waste composition vector, the industrial solid waste is associated with the process to obtain a composition-process matching weight matrix, which is the core step to realize intelligent matching of treatment process and subsequent cost optimization. After obtaining the solid waste composition vector composed of multiple substance contents, the system analyzes the differences in physical and chemical properties of different components, and combines the applicability and processing efficiency of various treatment processes (such as incineration, melting, sorting, solidification, etc.) to specific components to construct a multi-dimensional mapping relationship between components and processes. This process usually relies on a pre-established knowledge graph or rule engine to quantify the strength of the interaction between each component and all available processes, and to represent the matching priority and influence weight of each component on different processes in matrix form, thereby forming a composition-process matching weight matrix. For example, in the solid waste treatment scenario of the steel and iron industry, if the solid waste composition vector shows that it is rich in iron oxides and silicates, the system will give higher matching weights to high-temperature melting and magnetic separation processes, and give weights close to zero to irrelevant processes such as organic matter pyrolysis. This structured matrix expression not only clearly reflects the complex relationship between component characteristics and process adaptation, but also provides key input parameters for the subsequent global cost control model, making the entire treatment process more targeted and economical.

[0059] Step S5, based on the composition-process matching weight matrix, global cost control is performed to obtain a target cost treatment strategy scheme.

[0060] Specifically, based on the composition-process matching weight matrix, global cost control is performed to obtain a target cost treatment strategy scheme, which is the final key step in the method to realize economic optimization and intelligent decision-making. After constructing the composition-process matching weight matrix, the system takes the operating cost, energy consumption level, equipment depreciation, labor cost, etc. of various treatment processes as constraint conditions, combines the adaptation degree of each component to different processes in the weight matrix, establishes a multi-objective optimization model, and through linear programming or heuristic algorithm, all possible process combination paths are comprehensively evaluated and costed to find the optimal cost configuration scheme under the premise of meeting the treatment effect. For example, in the solid waste treatment process of the steel and iron industry, if the composition-process matching weight matrix shows that iron oxides are more suitable for high-temperature melting recovery, and silicate substances are more suitable for building material preparation, the system will dynamically generate a treatment strategy targeting the lowest cost or highest resource recovery value based on the unit treatment cost, resource recovery value, and emission cost of the two processes. This process not only realizes the whole-process closed-loop control from solid waste composition identification to treatment decision-making, but also significantly improves the economic efficiency and environmental sustainability of industrial solid waste treatment.

[0061] In specific embodiments, the multi-spectral scanning of the industrial solid waste by the multi-spectral imaging device to obtain solid waste multi-spectral image data comprises:

[0062] scanning the surface and internal structure of the industrial solid waste by a multispectral imaging device to obtain multispectral reflection and transmission image data;

[0063] correcting the spatial resolution in the multispectral reflection and transmission image data to generate corrected multispectral image data;

[0064] performing spectral feature enhancement on the corrected multispectral image data to obtain solid waste multispectral image data.

[0065] Specifically, the multi-spectral scanning of industrial solid waste by the multi-spectral imaging device to obtain solid waste multi-spectral image data is the starting link of information collection and data modeling in the entire industrial solid waste treatment cost control method, and is also a basic step for realizing subsequent component identification and process matching. This step not only includes multi-spectral imaging of the surface of industrial solid waste, but also further covers the collection of reflection and transmission information of its internal structure. Through multi-spectral imaging device scanning of solid waste samples at multiple discrete wavebands, reflectivity and transmissivity data at different wavelengths are obtained, thereby forming multi-band reflection and transmission image data containing spatial and spectral double information. For example, in the process of treating solid waste generated in the steel metallurgical industry, industrial waste slag often has a complex mineral composition and a heterogeneous structure. Traditional visible light imaging or single-band infrared imaging cannot fully reveal the internal material distribution characteristics. However, multi-spectral imaging devices can penetrate the waste slag through multiple wavebands (such as visible light, near-infrared, short-wave infrared, etc.) to capture the response differences at different electromagnetic wavebands, thereby obtaining more comprehensive material information. After obtaining the multi-band reflection and transmission image data, since the spatial resolution of different waveband images may not be consistent, the system will further correct the resolution of these image data to ensure that the waveband images are aligned in the spatial dimension and have a uniform pixel size, thereby generating corrected multi-spectral image data. This process usually uses image registration and interpolation algorithms to combine the optical parameters and imaging conditions of the device to perform geometric correction and spatial normalization of the images, avoiding errors in subsequent feature extraction caused by resolution differences. For example, during waste slag scanning, the imaging sensors of different wavebands may have different field angles or focal lengths, resulting in deformation or misalignment between images. After correction, the system can ensure that each pixel point corresponds to the same spatial position in all wavebands, providing an accurate spatial reference for subsequent spectral analysis. After completing the unification of spatial resolution, in order to further improve the distinguishability of the spectral characteristics of the materials in the image, the system will perform spectral feature enhancement processing on the corrected multi-spectral image data. This processing usually includes steps such as denoising, background subtraction, spectral normalization, and principal component analysis, aiming to highlight the differences in spectral dimension of different materials and improve the accuracy of subsequent component identification. For example, in the waste slag processing scenario, iron oxides and silicates may exhibit similar gray scale characteristics in the visible light waveband, but have significant reflectivity differences in the near-infrared waveband. By enhancing the spectral features, the distinguishability of these two types of materials in the image can be significantly improved, thereby improving the accuracy of component identification. Finally, after multi-band scanning, spatial correction, and spectral enhancement processing, the solid waste multi-spectral image data obtained can provide a high-quality data basis for subsequent waveband response feature extraction and component analysis, thereby supporting the intelligentization and cost optimization of the entire industrial solid waste treatment process.

[0066] In specific embodiments, the band response feature extraction on the solid waste multi-spectral image data to obtain the material spectral fingerprint feature comprises:

[0067] The inter-band correlation filtering is performed on the solid waste multi-spectral image data, and the filtered solid waste multi-spectral image data is subjected to spatial domain frequency decomposition to obtain spectral response data;

[0068] The local band feature in the spectral response data is extracted by an adaptive window sliding method to obtain a sequence of local band extreme points, and spectral absorption peak position identification is performed based on the sequence of local band extreme points to obtain characteristic peak position coordinates;

[0069] Based on the characteristic peak position coordinates, the band intervals of the spectral response data are divided to obtain a plurality of characteristic band subintervals, and the spectral reflectance gradient of the characteristic band subintervals is calculated to obtain a band gradient feature vector;

[0070] Spectral feature dimension reduction is performed based on the band gradient feature vector to obtain a low-dimensional spectral feature matrix, and eigenvalue decomposition is performed on the low-dimensional spectral feature matrix to obtain a material spectral fingerprint feature.

[0071] Specifically, the band response feature extraction of the solid waste multi-spectral image data to obtain the material spectral fingerprint feature is one of the key technical paths to realize the industrial solid waste component identification and subsequent process matching. Based on obtaining high-quality solid waste multi-spectral image data, the spectral fingerprint feature reflecting the essential properties of the material is extracted from the complex image information through multi-level data processing means, thereby providing high-precision and computable input parameters for building the component vector and establishing the process association. Specifically, in the solid waste multi-spectral image data after correction and enhancement, there may be high correlation or redundant information between different bands, so the system first filters the correlation between bands to remove noise interference and repetitive band content, and improves the data processing efficiency. Subsequently, the filtered solid waste multi-spectral image data is converted to the spatial domain frequency decomposition space, and the spatial-spectral structure of the image is decoupled by using wavelet transform or multi-scale analysis method, so as to separate the spectral response data at different frequency levels. For example, in the process of treating solid waste generated in the steel metallurgical industry, the waste slag contains various mineral components such as iron oxides and silicates, which have different reflection and absorption characteristics at different bands. By frequency decomposition of these data, the surface texture information and deep material information can be effectively distinguished, and the accuracy of feature extraction is improved. On this basis, the system uses an adaptive window sliding method to extract local band features from the spectral response data, that is, the size of the sliding window is dynamically adjusted in the neighborhood range of each pixel point, and the local extreme point sequence is identified in combination with the change trend of the spectral curve. These extreme points often correspond to the absorption or reflection peak of the material at a certain wavelength, and have strong physical meaning. The system further identifies the spectral absorption peak based on the local extreme point sequence of the band, and locates the key feature peak coordinates. For example, in the waste slag sample, if there is a significant absorption valley near a certain band, the peak may indicate the presence of iron oxide, and the reflection peak of another band may represent silicate minerals. Subsequently, the system divides the band interval according to the identified feature peak coordinates, forms several feature band subintervals with clear physical meaning, and calculates the change gradient of the spectral reflectance in each subinterval to generate a band gradient feature vector. This process not only preserves the details of the original spectrum, but also enhances the sensitivity to subtle component differences. In order to further reduce the feature dimension and improve the calculation efficiency, the system performs spectral feature dimension reduction operation on the obtained band gradient feature vector, usually using principal component analysis (PCA) or linear discriminant analysis (LDA) method to extract the most representative low-dimensional spectral feature matrix. Finally, by performing eigenvalue decomposition on the low-dimensional spectral feature matrix, the dominant eigenvalues and eigenvectors are extracted, thereby constructing the material spectral fingerprint feature that can uniquely identify a certain type of material.To sum up, this step realizes the conversion process from the original image data to the quantifiable spectral fingerprint feature through multiple links such as band correlation filtering, frequency decomposition, local feature extraction, peak position identification, band division, gradient calculation and feature dimension reduction, significantly improves the accuracy and stability of component identification, and provides solid data support for subsequent process matching and cost control.

[0072] In specific embodiments, the component determination of the industrial solid waste based on the substance spectral fingerprint feature comprises:

[0073] The substance spectral fingerprint feature is subjected to spectral line contrast calibration to obtain chemical element characteristic spectral data, and the substance components in the chemical element characteristic spectral data are analyzed to obtain an element composition distribution matrix;

[0074] The element composition distribution matrix is subjected to element quantitative analysis to obtain an element content distribution atlas, and the element content distribution atlas is subjected to spatial cluster analysis to obtain an element enrichment area parameter;

[0075] Based on the element enrichment area parameter, the compound structure of the industrial solid waste is deduced, a substance form combination list is generated, and the substance form combination list is subjected to stoichiometric balance verification to obtain a component allocation correction coefficient;

[0076] Through a substance balance constraint method, the component of the industrial solid waste is determined based on the component allocation correction coefficient to obtain a solid waste component vector.

[0077] Specifically, the composition determination of the industrial solid waste based on the substance spectral fingerprint feature is a key step to realize the conversion from high-dimensional spectral information to quantifiable and operable composition data. This process not only relies on the in-depth analysis of the substance spectral fingerprint feature, but also needs to combine chemometrics and image analysis technology to ensure that the final generated solid waste composition vector has high precision and interpretability. After obtaining the substance spectral fingerprint feature extracted by the previous step, the system first performs spectral line comparison calibration, that is, matches and compares the extracted spectral feature with the known substance spectrum in the standard spectral database to identify the corresponding chemical element spectral data. This process usually uses cross-correlation algorithm or spectral angle mapping (Spectral Angle Mapper, SAM) method to find the best matching point in the wavelength dimension, so as to identify the element types that may exist in the sample. For example, in the process of solid waste treatment in the steel and iron industry, the system can preliminarily judge whether elements such as iron (Fe), silicon (Si), calcium (Ca), and aluminum (Al) exist in the current solid waste sample by comparing the extracted spectral fingerprint with the standard spectrum of these elements, and further obtain the corresponding characteristic absorption or emission spectrum line position and intensity information, and then construct a chemical element spectral data set. Subsequently, the system analyzes the chemical element spectral data and identifies the differences in the spatial distribution of each element to form an element composition distribution matrix. This matrix describes the relative concentration changes of different elements in the image space, providing a basis for subsequent spatial clustering and quantitative analysis. On this basis, the system further quantitatively analyzes the element content distribution map, uses empirical models or machine learning regression methods to establish a mapping relationship between spectral reflectance or absorption intensity and the actual content of elements, and generates an element content distribution map. For example, for the iron element in a certain area, the system can estimate its mass percentage in the waste slag through its absorption depth at a certain wavelength, and then draw an iron element content distribution map to reflect the enrichment degree of iron in different regions. In order to more effectively identify regions with similar element composition, the system performs spatial clustering analysis on the above element content distribution map to identify element enrichment region parameters. This process usually uses unsupervised learning methods such as K-means clustering and DBSCAN to divide the entire image into several representative enrichment regions according to the spatial continuity and similarity of element content, and extracts the average content and coefficient of variation of the main elements in each region as the region parameters. For example, in the waste slag image, if a part shows high values of iron and oxygen elements, it may be an iron oxide enrichment area, while another part may show high values of silicon and aluminum, representing the presence of silicate minerals.Based on these element enrichment area parameters, the system further deduces the compound structure of the industrial solid waste, i.e. by the chemical affinity between known elements and the typical mineral phase combination rules, to infer the possible compound types such as oxides, carbonates, sulfides, etc., and generate a list of substance form combinations. For example, if the system detects that iron and oxygen elements coexist in a certain area with a proportion close to the theoretical molar ratio of Fe2O3 or Fe3O4, it will be classified into the corresponding iron oxide mineral category and recorded in the list of substance form combinations. In addition, the system will also consider the presence or absence of other elements and their relative proportions to exclude unreasonable mineral combinations. To further improve the accuracy of component identification, the system performs stoichiometric balance verification on the generated list of substance form combinations, calculates the component proportion correction coefficient by constraining the total mass conservation condition of each element in all fitted minerals. The core of this step is to establish a mass balance equation set, so that the theoretical total content of each element in all minerals is consistent with the actual measured value as much as possible, thereby correcting the deviation caused by spectral overlap, model error or background interference, etc. For example, when the system finds that a certain mineral combination causes the total amount of calcium element to exceed the measured value, it will automatically adjust the proportion of that mineral or introduce other calcium-containing minerals to meet the mass conservation condition. Finally, the system determines the industrial solid waste component composition that best fits the observation data by considering the component proportion correction coefficient and the stability of the mineral combination through the mass balance constraint method, and generates a solid waste component vector containing multiple compounds and their mass proportions. This component vector is not only an important input variable for subsequent process matching and cost control, but also constitutes the data core of the entire intelligent processing flow, realizing a complete closed loop from the original image to accurate component identification, and significantly improving the scientificity and economy of industrial solid waste resource processing.

[0078] In specific embodiments, the element composition distribution matrix is subjected to element quantitative analysis to obtain an element content distribution spectrum, including:

[0079] The element composition distribution matrix is subjected to atomic emission spectrum intensity correction to obtain an element characteristic spectrum line intensity table, and the element characteristic spectrum line intensity table is subjected to matrix effect compensation calculation to obtain an element sensitivity coefficient matrix;

[0080] The element sensitivity coefficient matrix is subjected to interference peak deconvolution by an element internal standard method to obtain an element pure spectrum, and the element pure spectrum is subjected to spectrum peak area integration to generate an element content calibration curve;

[0081] Based on the element content calibration curve, regional element concentration mapping is performed to obtain an element spatial distribution vector group, and the element spatial distribution vector group is subjected to element correlation clustering to generate an element paragenetic relationship spectrum;

[0082] Based on the element paragenetic relationship spectrum, element partition coefficient calculation is performed to obtain an element occurrence state mapping diagram, and chemical valence state balance constraint is performed on the element occurrence state mapping diagram to obtain an element content distribution spectrum.

[0083] Specifically, the element composition distribution matrix is subjected to element quantitative analysis to obtain an element content distribution map, which is an important link for changing from qualitative identification to quantitative evaluation and is also a core component of data analysis in the industrial solid waste composition determination process. Through in-depth processing of the element composition distribution matrix, combined with atomic emission spectrum analysis, matrix effect compensation, internal standard correction, spectral deconvolution and integration, spatial mapping, and chemical constraints, a spatial distribution map reflecting the actual concentration of each element in the solid waste sample is constructed. Specifically, after obtaining the element composition distribution matrix generated by the previous steps, the system first corrects the atomic emission spectrum intensity to eliminate measurement deviations caused by inconsistent instrument responses or environmental disturbances. This process is based on the known spectral characteristics of standard samples to normalize the original spectral signal, thereby obtaining an element characteristic spectral line intensity table containing the characteristic spectral line intensity of each element at different wavelengths. For example, during the treatment of solid waste generated in the steel metallurgical industry, if a certain area shows a strong iron element characteristic peak, the system will record its emission intensity value at a specific wavelength band and establish a preliminary intensity reference system accordingly. Subsequently, the system further compensates the element characteristic spectral line intensity table for matrix effect to solve the problem of the influence of other coexisting elements in the sample on the spectral signal of the target element. Matrix effect refers to the presence of non-target elements in the sample, which may change the excitation efficiency or spectral morphology of the target element, thereby affecting the accuracy of quantitative analysis. Therefore, the system introduces a reference sample library with known matrix composition, combines statistical methods such as multiple regression model or partial least squares (PLS), calculates the sensitivity variation law of each element under different matrix conditions, and constructs an element sensitivity coefficient matrix accordingly. For example, in a waste slag sample containing a large amount of silicon and calcium, the spectral signal of iron element may be inhibited. After matrix effect compensation, the system can correct this influence so that the final obtained element signal is closer to the true content. On this basis, the system uses the element internal standard method to perform interference peak deconvolution processing on the element sensitivity coefficient matrix, i.e., a stable element that is not affected by the elements to be measured is selected as an internal standard, and its spectral signal is used as a reference to normalize the spectral signals of other elements, thereby effectively separating the disturbed spectral peaks and extracting pure element spectra. For example, in a waste slag sample, aluminum is often chosen as an internal standard element because it has stable emission characteristics and is not easily affected by strong interactions with other elements in most cases. By calculating the signal ratio of iron, silicon, calcium, and other target elements to aluminum, background interference can be effectively removed, and spectral analysis accuracy can be improved. Subsequently, the system integrates the spectral peak area of the element pure spectrum, i.e., calculates the total area of each element characteristic peak by numerical integration method, and compares it with the known concentration in the standard sample to establish an element content calibration curve.This process usually involves polynomial fitting or nonlinear regression modeling to ensure that the relationship between spectral signal intensity and element concentration has good linearity and repeatability. For example, by experimentally determining the spectral signals of a series of waste residue samples with different iron contents, and plotting their relationship with iron content, an iron element content calibration curve for subsequent analysis can be formed. Next, the system maps the element concentration of each pixel in the entire image area based on the element content calibration curve, generating an element spatial distribution vector group that describes the specific concentration values of each element at each pixel point, providing basic data for subsequent spatial clustering analysis. In order to reveal the spatial distribution patterns of different elements and their potential paragenetic relationships, the system performs element correlation clustering analysis on the element spatial distribution vector group, identifies highly correlated element combinations in spatial distribution, and generates an element paragenetic relationship spectrum. For example, in waste residue images, iron and oxygen, silicon and aluminum often show high spatial correlation, indicating that they may coexist in iron oxide or silicate minerals. To further clarify the occurrence state of elements in mineral structures, the system performs element partition coefficient calculation based on the element paragenetic relationship spectrum, i.e., estimates the distribution proportion of each element in different mineral phases according to the thermodynamic equilibrium principle and mineral phase transition model, and generates an element occurrence state mapping diagram. For example, if iron is mainly distributed in magnetite and hematite, the system will calculate the proportion of iron in these two minerals, respectively. Finally, the system performs chemical valence state balance constraint on the element occurrence state mapping diagram to ensure the total mass conservation of all elements and comply with the charge neutrality principle, thereby generating the final element content distribution spectrum.

[0084] In specific embodiments, the process correlation of the industrial solid waste based on the solid waste composition vector obtains a composition-process matching weight matrix, including:

[0085] Physicochemical property analysis is performed on the solid waste composition vector to obtain a solid waste reaction activity parameter group, and thermodynamic stability analysis is performed on the industrial solid waste based on the solid waste reaction activity parameter group to obtain a solid waste thermodynamic stability matrix;

[0086] The conversion rate vector is obtained by conversion rate prediction of the industrial solid waste through the solid waste thermodynamic stability matrix, and the composition-process matching weight matrix is obtained by process correlation degree calculation based on the conversion rate vector.

[0087] Specifically, the process of obtaining the component-process matching weight matrix by correlating the industrial solid waste based on the solid waste component vector aims to determine the optimal matching relationship between different solid waste components and treatment processes by analyzing the physicochemical properties and reaction activity parameters of the industrial solid waste, combining thermodynamic stability analysis and conversion rate prediction. First, this process requires in-depth analysis of the physicochemical properties of the solid waste component vector. This step includes identifying and quantifying the physical properties (such as density, melting point, etc.) and chemical properties (such as acidity, alkalinity, redox properties, etc.) of each component, thereby obtaining a set of reaction activity parameters that reflect the behavior of these components under specific conditions. For example, when dealing with solid waste generated by the metallurgical industry, iron ore residues may contain incompletely reduced iron oxides and other impurity elements. The reaction activity parameters not only involve the reduction potential of iron oxides, but also include the existence form and reaction activity of other metal oxides or sulfides. Next, based on the solid waste reaction activity parameter set, the system will perform a thermodynamic stability analysis on the industrial solid waste to assess whether each component is stable or prone to chemical changes under different temperature, pressure, and environmental conditions. This thermodynamic stability analysis usually relies on standard enthalpy of formation, entropy change, and free energy change, etc. thermodynamic parameters, by calculating the change of free energy under different reaction paths, to predict which components are more likely to participate in chemical reactions and produce new substances. For example, under high temperature conditions, some insoluble compounds in low-grade ores may be converted into more soluble forms, thereby facilitating subsequent resource recovery and utilization. Through such analysis, a matrix describing the thermodynamic stability of each component in the solid waste under different process conditions can be constructed - the solid waste thermodynamic stability matrix. With this solid waste thermodynamic stability matrix, the next step is to predict the conversion rate of the industrial solid waste based on it to obtain the conversion rate vector. The conversion rate here refers to the speed at which a solid waste component changes from one state to another under specific process conditions. This step not only depends on the thermodynamic information obtained in the previous step, but also needs to consider kinetic factors such as reaction path, catalyst effect, etc. For example, in the hydrometallurgical process, in order to improve the leaching efficiency of certain metal ions, in addition to ensuring that the system has sufficient thermodynamic driving force, appropriate leaching agents and operating conditions need to be selected to accelerate the reaction rate. Therefore, based on known thermodynamic data and experimentally determined kinetic parameters, the conversion rate of different components under specified process conditions can be predicted, and the corresponding conversion rate vector can be formed. Finally, based on the conversion rate vector, the process correlation degree is calculated to obtain the component-process matching weight matrix. This step essentially evaluates the degree of adaptation between each component and various potential treatment processes by considering the physicochemical properties, thermodynamic stability, and conversion rate of the solid waste components.For example, for electronic waste containing a high proportion of heavy metal ions and organic pollutants, if the traditional incineration treatment method is adopted, the volatility of each component under high temperature conditions, the risk of toxic release and other factors need to be evaluated; if bioremediation technology is adopted, the microbial degradation capacity and environmental adaptability need to be concerned. By comparing the effects of different treatment methods on each component, a matching weight value for each component for a specific process can be assigned, and finally a component-process matching weight matrix that comprehensively reflects the correlation strength between all components and processes is formed. This matrix not only helps to optimize the treatment scheme of industrial solid waste, but also provides a scientific basis for maximizing the recycling of resources. In the whole process, whether it is the analysis of the physical and chemical properties of the solid waste component vector or the study of its thermodynamic stability and conversion rate, it is to more accurately understand the essential characteristics of solid waste in order to find the most suitable treatment process and ensure that the dual goals of environmental protection and efficient resource utilization are achieved.

[0088] In specific embodiments, the conversion rate prediction of the industrial solid waste by the solid waste thermodynamic stability matrix comprises:

[0089] The chemical equilibrium calculation of the solid waste thermodynamic stability matrix obtains a free energy change sequence, and the reaction path analysis of the free energy change sequence obtains a reaction path atlas;

[0090] Based on the reaction path atlas, the reaction kinetics parameter analysis of the industrial solid waste obtains a kinetics parameter group, and the temperature dependence analysis of the kinetics parameter group obtains an activation energy distribution spectrum;

[0091] The diffusion resistance calculation of the activation energy distribution spectrum obtains a diffusion limitation coefficient matrix, and the interface mass transfer effect analysis of the diffusion limitation coefficient matrix obtains a mass transfer resistance spectrum;

[0092] Based on the mass transfer resistance spectrum, the reaction rate estimation of the industrial solid waste obtains a rate control parameter, and the coupling effect analysis of the rate control parameter obtains a rate coupling matrix;

[0093] The rate limitation judgment of the rate coupling matrix obtains a rate limiting factor sequence, and the conversion rate calculation based on the rate limiting factor sequence obtains a conversion rate vector.

[0094] Specifically, first, after obtaining the solid waste thermodynamic stability matrix generated by the previous step, the system performs chemical equilibrium calculation on it to determine the possible chemical reaction paths and their corresponding free energy change sequences in the solid waste under different temperature, pressure and atmosphere conditions. This process is based on standard Gibbs free energy generation data, combined with actual operating environment parameters (such as pH value, oxidation-reduction potential, etc.), and uses Thermo-Calc or FactSage and other thermodynamic software platforms for multiphase equilibrium simulation to obtain the ΔG value sequence of each possible reaction. For example, in the process of treating solid waste generated in the steel and iron metallurgical industry, if the waste slag contains incompletely reduced FeO, CaO and other metal oxides, under high-temperature reducing atmosphere, these oxides will undergo the following reaction: FeO + CO → Fe + CO2, at this time the system will judge whether it has the tendency to proceed spontaneously according to the ΔG value of the reaction, and include it in the subsequent kinetic analysis range. Next, the system performs reaction path analysis on the free energy change sequence to identify the dominant reaction path with the lowest energy and the highest possibility of occurring, and then constructs a reaction path map. This map not only includes the main reaction path, but also covers possible side reactions and the generation path of intermediate products, providing structured support for in-depth understanding of the chemical conversion mechanism in complex systems. For example, in the high-temperature melting treatment process of waste slag, in addition to the main metal reduction reaction, there may also be silicate decomposition, sulfide volatilization and other secondary paths, and the system will sort them according to the thermodynamic stability sequence to clarify the competition relationship between the paths. Subsequently, based on the reaction path map, the system further analyzes the reaction kinetics parameters of the industrial solid waste, that is, based on the known reaction path, the Arrhenius equation and related experimental data are introduced to estimate the rate constant, reaction order and other key kinetic parameters of each reaction, and finally form a kinetic parameter group. This step usually needs to rely on a kinetic database (such as CHEMKIN) or a machine learning regression model, combined with the measured reaction rate data, to fit the kinetic expression suitable for specific process conditions. For example, in the process of high-temperature gasification of waste slag, the oxidation reaction rate of carbon has an exponential relationship with temperature, and the system can describe its temperature variation trend through the kinetic parameter group. In order to more accurately reflect the influence of actual process conditions, the system further analyzes the temperature dependence of the kinetic parameter group and extracts the activation energy distribution spectrum of each reaction. As an important indicator of measuring the ease of reaction initiation, activation energy can effectively distinguish the activity differences of different reaction paths in different temperature intervals. For example, some high-activation-energy reactions are difficult to proceed at low temperatures, but show extremely high reaction rates at high temperatures, and the system can thus determine which process conditions are more conducive to the generation of target products. On this basis, the system calculates the diffusion resistance of the activation energy distribution spectrum and constructs a diffusion limitation coefficient matrix considering the problem of limited mass transfer within solid waste particles.Due to the heterogeneous and dense structure of most industrial solid wastes, the diffusion rate of reactants or products in the solid phase often becomes an important factor restricting the overall reaction rate. Therefore, the system adopts Fick diffusion model or random pore model, combined with physical parameters such as particle size, porosity, and specific surface area, to estimate the diffusion resistance in each reaction path, and incorporates it into the overall rate model. Subsequently, the system further analyzes the interfacial mass transfer effect of the diffusion limitation coefficient matrix, i.e. considers the mass transfer problem when the reaction occurs on the solid-liquid, solid-gas or multi-phase interface, identifies the additional mass transfer resistance caused by poor interfacial contact or uneven concentration gradient, and generates a mass transfer resistance spectrum. For example, in the process of treating heavy metal-containing waste residues by wet leaching, if H+ ions in the liquid phase cannot quickly diffuse to the mineral surface, it will cause a decrease in local reaction rate, and the system can quantify such effects through the mass transfer resistance spectrum. Then, the system estimates the reaction rate of the industrial solid waste based on the mass transfer resistance spectrum, considering factors such as thermodynamic driving force, kinetic rate constant, diffusion limitation, and mass transfer resistance, to establish a complete rate control parameter set. This parameter set not only includes the intrinsic rate of each reaction, but also includes the macroscopic rate controlled by external conditions, providing a basis for subsequent rate coupling modeling. To further reveal the interaction between multiple reaction paths, the system analyzes the coupling effect of the rate control parameters, identifies the synergistic or inhibitory effect between coexisting reactions due to competition for resources, changes in local environment, or generation of inhibitory intermediates, and finally generates a rate coupling matrix. For example, in a certain treatment process, the reduction reaction of iron oxides may release a large amount of heat, promoting the decomposition of other minerals, and this positive coupling effect will be reflected in the rate coupling matrix. Finally, the system judges the rate limitation of the rate coupling matrix and identifies the rate-limiting factor sequence that plays a decisive role in the entire reaction system. These factors may be the low rate constant of a certain reaction, high diffusion resistance, or severe interfacial mass transfer limitation. Based on the rate-limiting factor sequence, the system finally calculates the conversion rate and outputs the conversion rate vector containing the conversion rate information of each component under different process conditions. In summary, this step realizes the comprehensive prediction of the conversion behavior of industrial solid waste under complex process conditions through the integration of multiple levels of technologies such as chemical equilibrium calculation, reaction path analysis, kinetic parameter analysis, temperature dependence modeling, diffusion resistance evaluation, interfacial mass transfer effect analysis, rate estimation and coupling effect modeling, rate-limiting factor identification, etc. This process not only improves the scientificity of process selection, but also provides a solid data support for subsequent cost optimization and resource allocation, ensuring that the entire industrial solid waste treatment process is both efficient and economical.

[0095] In specific embodiments, the global cost calculation control based on the ingredient-process matching weight matrix to obtain a target cost treatment strategy scheme includes:

[0096] dynamically planning and decomposing the ingredient-process matching weight matrix to obtain a process unit cost constraint condition set, and performing process scheduling optimization based on the process unit cost constraint condition set to obtain a processing process timing arrangement table;

[0097] performing resource allocation calculation on the processing process timing arrangement table through multi-stage cost accumulation to obtain a resource scheduling strategy matrix, and performing boundary constraint verification on the resource scheduling strategy matrix to obtain a process operation parameter configuration set;

[0098] performing cost control index decomposition based on the process operation parameter configuration set to obtain a target cost processing strategy scheme.

[0099] Specifically, after obtaining the ingredient-process matching weight matrix generated by the previous step, the system first performs dynamic programming decomposition, i.e., divides the entire processing process into several process units, and sets the operation priority and cost constraint conditions of each process unit according to the adaptation degree of each ingredient to different processes, thereby generating a set of process unit cost constraint conditions. This process usually uses dynamic programming algorithm or mixed integer linear programming (MILP) modeling method, which regards process selection and operation sequence as variables, takes minimizing total processing cost as objective function, and combines multi-dimensional cost parameters such as energy consumption, labor, equipment depreciation, etc. to perform multi-stage path optimization. For example, in the solid waste treatment process generated in the steel metallurgical industry, if a batch of waste slag is rich in iron oxides, then the high-temperature reduction smelting process may be given a higher priority, while other low-value components such as silicates may enter the building material utilization path. Through dynamic programming decomposition, the application range and cost boundary of each process unit can be clearly identified. Subsequently, the system performs process scheduling optimization based on the set of process unit cost constraint conditions, i.e., reasonably arranges the execution order and parallel relationship of each process unit under the premise of meeting processing efficiency and resource availability, and generates a processing process timing table. This step usually relies on scheduling optimization algorithms such as genetic algorithm, ant colony algorithm or heuristic rule reasoning to ensure that the entire processing flow is feasible and efficient in the time dimension. For example, in actual operation, there are dependency relationships between processes such as magnetic separation, crushing and screening, and high-temperature melting, which must be executed in a certain logical order, while also taking into account equipment load balancing and energy consumption optimization. On this basis, the system further performs resource allocation calculation on the processing process timing table through a multi-stage cost accumulation mechanism, i.e., according to the time node, required material quantity, and equipment demand of each process, deduces the corresponding resource scheduling strategy matrix. This matrix covers the allocation plan of manpower, electricity, fuel, auxiliary materials, and other resources, providing data support for subsequent operation control. For example, in the high-temperature treatment process of waste slag, the system will predict the required power supply curve based on the operation cycle and power demand of the melting furnace, and develop a reasonable operation personnel configuration scheme combined with the personnel shift system, thereby forming a complete resource scheduling strategy matrix. To ensure that the developed resource scheduling strategy meets the actual operation conditions, the system also needs to perform boundary constraint checking on the resource scheduling strategy matrix, including hard constraint conditions such as device capacity limit, energy consumption upper limit, and environmental emission standard, as well as soft constraint factors such as process window period and raw material inventory level. By verifying these constraint conditions one by one, the system can identify potential risk points and make adjustments, ultimately generating a set of process operation parameter configurations that meet the requirements of on-site operation. For example, if the natural gas supply required by a process exceeds the current gas tank capacity, the system will automatically adjust the start time of that process or introduce an alternative energy scheme to ensure the stability and safety of the overall process.Finally, the system decomposes the cost control index based on the process operation parameter configuration set, that is, the total cost target is refined into various process units, process stages and resource elements, forming a multi-dimensional cost structure system covering direct costs (such as raw materials, energy consumption), indirect costs (such as equipment maintenance, labor management) and environmental costs (such as carbon emissions, wastewater treatment). Through the establishment of a cost feedback mechanism and an optimization iteration model, the system can dynamically adjust the resource allocation strategy and the process operation parameter, constantly approaching the optimal cost control target, and finally outputting a target cost treatment strategy scheme containing the treatment path, resource allocation, cost composition, etc. In summary, through the synergistic effect of multiple technical modules such as dynamic planning decomposition, process scheduling optimization, resource allocation calculation, boundary constraint verification and cost index decomposition, this step realizes complete closed-loop control from process matching relationship to economic decision-making, not only improving the intelligent level of industrial solid waste treatment process, but also providing scientific basis and technical support for enterprises to seek the best balance point between resource utilization and cost control.

[0100] The cost control method for industrial solid waste treatment in the embodiments of the present application is described above, and the cost control system for industrial solid waste treatment in the embodiments of the present application is described below. Please refer to Figure 2 The cost control system for industrial solid waste treatment in the embodiments of the present application includes one embodiment:

[0101] The scanning module 21 is configured to perform multi-spectral scanning on the industrial solid waste by a multi-spectral imaging device to obtain solid waste multi-spectral image data.

[0102] The extraction module 22 is configured to extract a material spectral fingerprint feature from the solid waste multi-spectral image data.

[0103] The determination module 23 is configured to determine the composition of the industrial solid waste based on the material spectral fingerprint feature to obtain a solid waste composition vector.

[0104] The association module 24 is configured to associate the industrial solid waste with a process based on the solid waste composition vector to obtain a composition-process matching weight matrix.

[0105] The control module 25 is configured to perform global cost control based on the composition-process matching weight matrix to obtain a target cost treatment strategy scheme.

[0106] In this embodiment, the specific implementation of each unit in the above system embodiment is described in the above method embodiment, which will not be repeated here.

[0107] Referring to Figure 3 In the embodiments of the present application, a computer device is also provided, and the internal structure of the computer device can be as Figure 3The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store corresponding data in the embodiment. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement the above method.

[0108] Those skilled in the art can understand that, Figure 3 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied.

[0109] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program. The computer program is executed by the processor to implement the above method. It can be understood that the computer readable storage medium in the embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0110] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiment method. Any reference to the memory, storage, database or other medium provided by the present application and used in the embodiment can include non-volatile and / or volatile memory. The non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. The volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM, etc.

[0111] It is to be understood that the terminology "including", "comprising", or any other variation thereof, is intended to cover a non-exclusive inclusion such that process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a... " does not, without more constraints, exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0112] The above description is merely the preferred embodiments of the present application, and is not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made according to the content of the present application specification and drawings, or directly or indirectly applied to other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A cost control method for industrial solid waste treatment, characterized in that, Includes the following steps: Multispectral imaging equipment was used to perform multispectral scanning on industrial solid waste to obtain multispectral image data of the solid waste; Band response features are extracted from the solid waste multispectral image data to obtain the material spectral fingerprint features; The composition of the industrial solid waste is determined based on the spectral fingerprint characteristics of the material, and a solid waste composition vector is obtained. Based on the solid waste component vector, the industrial solid waste is correlated with the process to obtain the component-process matching weight matrix; Global cost control is performed based on the component-process matching weight matrix to obtain a target cost handling strategy. The process association of the industrial solid waste based on the solid waste component vector to obtain the component-process matching weight matrix includes: The physicochemical properties of the solid waste component vector are analyzed to obtain a set of solid waste reactivity parameters. Based on the set of solid waste reactivity parameters, the thermodynamic stability of the industrial solid waste is analyzed to obtain a solid waste thermodynamic stability matrix. The conversion rate of the industrial solid waste is predicted by the solid waste thermodynamic stability matrix to obtain a conversion rate vector, and the process correlation degree is calculated based on the conversion rate vector to obtain a component-process matching weight matrix. The global cost control based on the component-process matching weight matrix, to obtain the target cost handling strategy, includes: The component-process matching weight matrix is ​​dynamically decomposed to obtain the process unit cost constraint set, and the process scheduling is optimized based on the process unit cost constraint set to obtain the processing process sequence arrangement table. By accumulating costs in multiple stages, resource allocation calculations are performed on the processing sequence schedule table to obtain a resource scheduling strategy matrix. Boundary constraint verification is then performed on the resource scheduling strategy matrix to obtain a process operation parameter configuration set. Based on the set of process operation parameters, cost control indicators are decomposed to obtain a target cost handling strategy.

2. The cost control method for industrial solid waste treatment according to claim 1, characterized in that, The process of performing multispectral scanning of industrial solid waste using a multispectral imaging device to obtain multispectral image data of the solid waste includes: The surface and internal structure of the industrial solid waste were scanned using a multispectral imaging device to obtain multi-band reflection and transmission image data. Correct the spatial resolution in the multi-band reflection and transmission image data to generate corrected multispectral image data; The spectral features of the corrected multispectral image data are enhanced to obtain solid waste multispectral image data.

3. The cost control method for industrial solid waste treatment according to claim 1, characterized in that, The step of extracting band response features from the solid waste multispectral image data to obtain material spectral fingerprint features includes: Inter-band correlation filtering is performed on the solid waste multispectral image data, and spatial domain frequency decomposition is performed on the filtered solid waste multispectral image data to obtain spectral response data. Local band features in the spectral response data are extracted using an adaptive window sliding method to obtain a sequence of local extreme points in the bands. Based on the sequence of local extreme points in the bands, the spectral absorption peak positions are identified to obtain the coordinates of the characteristic peak positions. Based on the characteristic peak position coordinates, the band intervals of the spectral response data are divided to obtain several characteristic band sub-intervals, and the spectral reflectance gradient of the characteristic band sub-intervals is calculated to obtain the band gradient feature vector. Based on the band gradient feature vector, the spectral feature dimension is reduced to obtain a low-dimensional spectral feature matrix. Then, the low-dimensional spectral feature matrix is ​​decomposed into eigenvalues ​​to obtain the spectral fingerprint features of the material.

4. The cost control method for industrial solid waste treatment according to claim 1, characterized in that, The process of determining the composition of the industrial solid waste based on the spectral fingerprint features of the material to obtain a solid waste composition vector includes: The spectral fingerprint features of the substance are compared and calibrated to obtain the characteristic spectral data of chemical elements, and the substance components in the characteristic spectral data of chemical elements are analyzed to obtain the elemental composition distribution matrix. The elements are arranged into a distribution matrix for quantitative analysis to obtain an element content distribution map. Spatial cluster analysis is then performed on the element content distribution map to obtain element enrichment region parameters. Based on the parameters of the enriched element regions, the compound structure of the industrial solid waste is deduced, a list of material form combinations is generated, and the list of material form combinations is stoichiometrically verified to obtain the component ratio correction coefficient. The composition of the industrial solid waste is determined by the material balance constraint method based on the component ratio correction coefficient, and a solid waste composition vector is obtained.

5. The cost control method for industrial solid waste treatment according to claim 4, characterized in that, The step of forming a distribution matrix of the elements and performing quantitative elemental analysis to obtain an elemental content distribution map includes: The elemental composition distribution matrix is ​​corrected by atomic emission spectrum intensity to obtain an elemental characteristic spectral line intensity table, and matrix effect compensation calculation is performed on the elemental characteristic spectral line intensity table to obtain an elemental sensitivity coefficient matrix. The element sensitivity coefficient matrix is ​​de-overlapped by the element internal standard method to obtain the element pure spectrum, and the element content calibration curve is generated by integrating the spectral peak area of ​​the element pure spectrum. Based on the element content calibration curve, regional element concentration mapping is performed to obtain a set of element spatial distribution vectors, and element correlation clustering is performed on the set of element spatial distribution vectors to generate an element co-occurrence relationship spectrum. Based on the element symbiotic relationship spectrum, the element partition coefficient is calculated to obtain the element occurrence state mapping map, and the chemical valence equilibrium constraint is applied to the element occurrence state mapping map to obtain the element content distribution map.

6. A cost control system for industrial solid waste treatment, characterized in that, include: The scanning module is used to perform multispectral scanning of industrial solid waste using a multispectral imaging device to obtain multispectral image data of the solid waste. The extraction module is used to extract band response features from the solid waste multispectral image data to obtain the material spectral fingerprint features; The determination module is used to determine the composition of the industrial solid waste based on the spectral fingerprint features of the substance, and obtain a solid waste composition vector. The association module is used to perform process association on the industrial solid waste based on the solid waste component vector to obtain a component-process matching weight matrix. The control module is used to perform global cost control based on the component-process matching weight matrix to obtain a target cost processing strategy. The process association of the industrial solid waste based on the solid waste component vector to obtain the component-process matching weight matrix includes: The physicochemical properties of the solid waste component vector are analyzed to obtain a set of solid waste reactivity parameters. Based on the set of solid waste reactivity parameters, the thermodynamic stability of the industrial solid waste is analyzed to obtain a solid waste thermodynamic stability matrix. The conversion rate of the industrial solid waste is predicted by the solid waste thermodynamic stability matrix to obtain a conversion rate vector, and the process correlation degree is calculated based on the conversion rate vector to obtain a component-process matching weight matrix. The global cost control based on the component-process matching weight matrix, to obtain the target cost handling strategy, includes: The component-process matching weight matrix is ​​dynamically decomposed to obtain the process unit cost constraint set, and the process scheduling is optimized based on the process unit cost constraint set to obtain the processing process sequence arrangement table. By accumulating costs in multiple stages, resource allocation calculations are performed on the processing sequence schedule table to obtain a resource scheduling strategy matrix. Boundary constraint verification is then performed on the resource scheduling strategy matrix to obtain a process operation parameter configuration set. Based on the set of process operation parameters, cost control indicators are decomposed to obtain a target cost handling strategy.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images