Method, device and equipment for predicting chemical component content of lead-barium glass before weathering
By using the K-means clustering algorithm and the multidimensional vector mapping state transformation equation, the problem of relying on experience to classify the weathering and non-weathering points of ancient lead-barium glass artifacts was solved, and the accurate prediction of the chemical composition content before weathering was achieved, thus improving the prediction accuracy.
Patent Information
- Application Number
- CN202310722363.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-16
- Publication Date
- 2025-12-16
- Estimated Expiration
- 2043-06-16
AI Technical Summary
In existing technologies, the classification of weathering and non-weathering points of ancient lead-barium glass artifacts relies on the experience of technicians, resulting in highly subjective and low-precision classification results, which affects the accuracy of predicting the chemical composition content before weathering.
The K-means clustering algorithm is used to cluster the lead-barium glass dataset, and a state transformation equation for a multidimensional vector mapping is constructed. By minimizing the mean square error, the state transformation equation is optimized to achieve accurate classification of weathered and unweathered points and prediction of chemical composition content.
It enables accurate and objective prediction of the chemical composition of ancient lead-barium glass artifacts before weathering, improving classification and prediction accuracy.
Smart Images
Figure CN116796215B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of chemical composition content prediction technology, and in particular to a method, apparatus and equipment for predicting the chemical composition content of lead-barium glass before weathering. Background Technology
[0002] China has a long history of glassmaking. After glass was introduced to China from West Asia and Egypt via the Silk Road, Chinese workers absorbed the techniques and used locally sourced materials to create ancient glass artifacts. Lead-barium glass is a special type of glass characterized by its primary content of lead (Pb) and barium (Ba). Lead-barium glass was a major material in the production of ancient glass artifacts. These artifacts were susceptible to weathering due to burial conditions, leading to significant exchange of elements between the glass and the surrounding environment. This altered the proportions of their chemical composition, affecting the analysis of their original chemical properties.
[0003] Currently, the general approach involves statistically analyzing the chemical composition data of weathered and unweathered areas of ancient lead-barium glass artifacts to establish a model relating component content to the degree of weathering. This model is then used to predict the chemical composition content before weathering, based on the weathering data of the weathered areas. However, the distinction between weathered and unweathered areas is currently made by technicians observing the appearance and structure of ancient lead-barium glass artifacts. The accuracy of this classification depends heavily on the experience of these technicians. Furthermore, since there is no clear boundary between weathered and unweathered areas—instead of a transitional zone—the subjective nature of this manual classification leads to low accuracy, thus affecting the prediction accuracy of the chemical composition content before weathering. Summary of the Invention
[0004] Therefore, it is necessary to provide a method, apparatus, and equipment for predicting the chemical composition content of lead-barium glass before weathering, in order to address the above-mentioned technical problems.
[0005] A method for predicting the chemical composition of lead-barium glass before weathering, the method comprising:
[0006] The chemical composition characteristics of ancient lead-barium glass artifacts were sampled and selectively labeled to obtain a lead-barium glass dataset; the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points.
[0007] The special dataset of lead-barium glass is clustered according to the K-means clustering algorithm to obtain the weathered and unweathered classifiers of lead-barium glass. The ordinary dataset of lead-barium glass is then input into the weathered and unweathered classifiers of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0008] A state transformation equation based on weathered point data and unweathered point data is constructed using a multidimensional vector mapping. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathered point data is then input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0009] In one embodiment, the chemical composition characteristics of ancient lead-barium glass artifacts are sampled and selectively labeled to obtain a lead-barium glass dataset, including:
[0010] The chemical composition characteristics of ancient lead-barium glass artifacts were sampled to obtain the types and contents of chemical components at each sampling point;
[0011] Sampling points are selectively labeled based on their weathering degree, resulting in labeled and unlabeled sample points. A special dataset for lead-barium glass is constructed based on the types and contents of chemical components at the labeled sample points, while a general dataset for lead-barium glass is constructed based on the types and contents of chemical components at the unlabeled sample points. The labeled sample points include labeled weathered points and labeled unweathered points.
[0012] In one embodiment, the chemical components at each sampling point include silicon dioxide, sulfur dioxide, phosphorus pentoxide, barium oxide, calcium oxide, potassium oxide, aluminum oxide, magnesium oxide, sodium oxide, lead oxide, strontium oxide, iron oxide, copper oxide, and tin oxide.
[0013] In one embodiment, after obtaining the lead-barium glass dataset, the method further includes:
[0014] Missing chemical composition values in the lead-barium glass dataset were padded with zeros. The proportion of different chemical compositions at each sampling point in the lead-barium glass dataset was calculated. The proportions of chemical compositions at the same sampling point were accumulated, and sampling points with proportions between 85% and 105% were considered as valid sampling points. By statistically analyzing the types and contents of chemical compositions at the valid sampling points, the quality-controlled lead-barium glass dataset was obtained.
[0015] In one embodiment, a special dataset of lead-barium glass is clustered using the K-means clustering algorithm to obtain a classifier for weathered and unweathered lead-barium glass, including:
[0016] Two labeled sample points in a special dataset of lead-barium glass are randomly selected as centroids. By calculating the distance between each labeled sample point and the two centroids, each labeled sample point is clustered into the cluster to which the nearest centroid belongs.
[0017] The average value of all labeled sample points in the cluster obtained by clustering is used as the new centroid, and the position of the centroid is updated.
[0018] Based on the new centroid, the labeled sample points in the lead-barium glass special dataset are repeatedly clustered until the centroid position no longer changes. Then, the clustering stops and the weathered and unweathered classifiers for lead-barium glass are output.
[0019] In one embodiment, a state transformation equation based on weathered point data and unweathered point data is constructed using a multi-dimensional vector mapping. The state transformation equation is then optimized by minimizing its mean square error to obtain the optimized state transformation equation, which includes:
[0020] Based on the weathering point data X = (x1, x2, x3, ..., x...), n ) and unweathered point data Y = (y1, y2, y3, ..., y n Construct the state transformation equation for the multidimensional vector mapping, expressed as G(x) i ) = w i x i +b i =y i ; where x i y represents the content of the i-th chemical component in the weathering point data. i w represents the content of the i-th chemical component in the unweathered point data. i and b i These represent the first and second parameters of the state transformation equation, respectively, and n = 14 represents the total number of chemical components.
[0021] The first and second parameters of the state transformation equation are optimized by minimizing the mean square error of the state transformation equation, resulting in the optimized state transformation equation.
[0022] In one embodiment, the first and second parameters of the state transformation equation are optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation, including:
[0023] The mean square error of the state transformation equation is expressed as:
[0024]
[0025] in, This represents the content of the i-th chemical component at the j-th unweathered point in the unweathered point data. This represents the content of the i-th chemical component at the j-th weathering point in the weathering point data, where m represents the total number of j;
[0026] By minimizing For the first parameter w of the state transformation equation i Second parameter b iOptimization is performed to obtain first and second optimization parameters. Based on these parameters, the optimized state transition equation is constructed, where the first and second optimization parameters are respectively expressed as...
[0027]
[0028]
[0029] Among them, W i B represents the first optimization parameter. i This represents the second optimization parameter. express The mean.
[0030] A device for predicting the chemical composition of lead-barium glass before weathering, the device comprising:
[0031] The data sampling module is used to sample and selectively label the chemical composition characteristics of ancient lead-barium glass artifacts to obtain a lead-barium glass dataset. The lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points.
[0032] The data clustering module is used to cluster the special dataset of lead-barium glass according to the K-means clustering algorithm to obtain the weathered and unweathered classifiers of lead-barium glass. The ordinary dataset of lead-barium glass is input into the weathered and unweathered classifiers of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0033] The prediction module is used to construct a state transformation equation for a multi-dimensional vector mapping based on weathered point data and unweathered point data. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathered point data is then input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0034] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:
[0035] The chemical composition characteristics of ancient lead-barium glass artifacts were sampled and selectively labeled to obtain a lead-barium glass dataset; the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points.
[0036] The special dataset of lead-barium glass is clustered according to the K-means clustering algorithm to obtain the weathered and unweathered classifiers of lead-barium glass. The ordinary dataset of lead-barium glass is then input into the weathered and unweathered classifiers of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0037] A state transformation equation based on weathered point data and unweathered point data is constructed using a multidimensional vector mapping. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathered point data is then input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0038] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:
[0039] The chemical composition characteristics of ancient lead-barium glass artifacts were sampled and selectively labeled to obtain a lead-barium glass dataset; the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points.
[0040] The special dataset of lead-barium glass is clustered according to the K-means clustering algorithm to obtain the weathered and unweathered classifiers of lead-barium glass. The ordinary dataset of lead-barium glass is then input into the weathered and unweathered classifiers of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0041] A state transformation equation based on weathered point data and unweathered point data is constructed using a multidimensional vector mapping. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathered point data is then input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0042] The aforementioned method, apparatus, and equipment for predicting the chemical composition content of lead-barium glass before weathering involves sampling and selectively labeling the chemical composition characteristics of ancient lead-barium glass artifacts to obtain a lead-barium glass dataset including a special dataset and a general dataset. The special dataset is then clustered using the K-means clustering algorithm to obtain a classifier for weathered and unweathered lead-barium glass. The general dataset is input into the classifier for classification to obtain weathered and unweathered point data. A state transformation equation based on the weathered and unweathered point data is constructed using a multi-dimensional vector mapping. This equation is then optimized by minimizing the mean square error of the state transformation equation to obtain an optimized state transformation equation. The weathered point data is then input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering. This method can accurately and objectively cluster weathered and unweathered points on ancient lead-barium glass artifacts using the K-means clustering algorithm. Furthermore, it can construct a state transformation equation for a multidimensional vector mapping based on the corresponding weathered and unweathered point data, thereby enabling accurate prediction of the chemical composition content of the weathered points before weathering. Attached Figure Description
[0043] Figure 1 This is a flowchart illustrating a method for predicting the chemical composition of lead-barium glass before weathering in one embodiment.
[0044] Figure 2 This is a structural block diagram of a device for predicting the chemical composition of lead-barium glass before weathering, as shown in one embodiment.
[0045] Figure 3 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation
[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0047] In one embodiment, such as Figure 1 As shown, a method for predicting the chemical composition of lead-barium glass before weathering is provided, including the following steps:
[0048] Step S1: Sample and selectively label the chemical composition characteristics of ancient lead-barium glass artifacts to obtain a lead-barium glass dataset; wherein, the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points.
[0049] Step S2: Cluster the special dataset of lead-barium glass according to the K-means clustering algorithm to obtain the weathered and unweathered classifiers of lead-barium glass. Input the ordinary dataset of lead-barium glass into the weathered and unweathered classifiers of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0050] As can be understood, the K-means clustering algorithm achieves clustering by finding the centroid of each cluster as a prototype and grouping the points adjacent to each centroid into different clusters. It is suitable for special datasets of lead-barium glass containing labeled sample points, where the clusters are distinct.
[0051] Step S3: Construct a state transformation equation for a multidimensional vector mapping based on weathered point data and unweathered point data. Optimize the state transformation equation by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. Input the weathered point data into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0052] It is understandable that the state transformation equation of multidimensional vector mapping uses the multidimensional vector in one state to predict the multidimensional vector in another state, and multidimensional vector mapping is used to predict the chemical composition content of the weathering point before weathering.
[0053] In one embodiment, the chemical composition characteristics of ancient lead-barium glass artifacts are sampled and selectively labeled to obtain a lead-barium glass dataset, including:
[0054] The chemical composition characteristics of ancient lead-barium glass artifacts were sampled to obtain the types and contents of chemical components at each sampling point. The types of chemical components included silicon dioxide, sulfur dioxide, phosphorus pentoxide, barium oxide, calcium oxide, potassium oxide, aluminum oxide, magnesium oxide, sodium oxide, lead oxide, strontium oxide, iron oxide, copper oxide, and tin oxide.
[0055] Sampling points are selectively labeled based on their weathering degree, resulting in labeled and unlabeled sample points. A special dataset for lead-barium glass is constructed based on the types and contents of chemical components at the labeled sample points, while a general dataset for lead-barium glass is constructed based on the types and contents of chemical components at the unlabeled sample points. The labeled sample points include labeled weathered points and labeled unweathered points.
[0056] It is understandable that the labeled weathered points and labeled unweathered points in the labeled sample points are severely weathered points and completely unweathered points, respectively. The data characteristics are distinct and suitable for K-means clustering algorithm to cluster, and generate a classifier to distinguish between unlabeled weathered points and unweathered points in the unlabeled sample points.
[0057] In one embodiment, after obtaining the lead-barium glass dataset, the method further includes:
[0058] Missing chemical composition values in the lead-barium glass dataset were padded with zeros. The proportion of different chemical compositions at each sampling point in the lead-barium glass dataset was calculated. The proportions of chemical compositions at the same sampling point were accumulated, and sampling points with proportions between 85% and 105% were considered as valid sampling points. By statistically analyzing the types and contents of chemical compositions at the valid sampling points, the quality-controlled lead-barium glass dataset was obtained.
[0059] It is understandable that by imputing missing values and removing outliers from the lead-barium glass dataset, the data quality was further improved.
[0060] In one embodiment, a special dataset of lead-barium glass is clustered using the K-means clustering algorithm to obtain a classifier for weathered and unweathered lead-barium glass, including:
[0061] Two labeled sample points in a special dataset of lead-barium glass are randomly selected as centroids. By calculating the distance between each labeled sample point and the two centroids, each labeled sample point is clustered into the cluster to which the nearest centroid belongs.
[0062] The average value of all labeled sample points in the cluster obtained by clustering is used as the new centroid, and the position of the centroid is updated.
[0063] Based on the new centroid, the labeled sample points in the lead-barium glass special dataset are repeatedly clustered until the centroid position no longer changes. Then, the clustering stops and the weathered and unweathered classifiers for lead-barium glass are output.
[0064] In one embodiment, a state transformation equation based on weathered point data and unweathered point data is constructed using a multi-dimensional vector mapping. The state transformation equation is then optimized by minimizing its mean square error to obtain the optimized state transformation equation, which includes:
[0065] Based on the weathering point data X = (x1, x2, x3, ..., x...), n ) and unweathered point data Y = (y1, y2, y3, ..., y n Construct the state transformation equation for the multidimensional vector mapping, expressed as G(x) i ) = w i x i +b i =y i ; where x i y represents the content of the i-th chemical component in the weathering point data. i w represents the content of the i-th chemical component in the unweathered point data. i and b iThese represent the first and second parameters of the state transformation equation, respectively, and n = 14 represents the total number of chemical components.
[0066] The first and second parameters of the state transformation equation are optimized by minimizing the mean square error of the state transformation equation, resulting in the optimized state transformation equation.
[0067] In one embodiment, the first and second parameters of the state transformation equation are optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation, including:
[0068] The mean square error of the state transformation equation is expressed as:
[0069]
[0070] in, This represents the content of the i-th chemical component at the j-th unweathered point in the unweathered point data. This represents the content of the i-th chemical component at the j-th weathering point in the weathering point data, where m represents the total number of j;
[0071] By minimizing For the first parameter w of the state transformation equation i Second parameter b i Optimization is performed to obtain the first optimization parameter and the second optimization parameter. Based on the first optimization parameter and the second optimization parameter, the optimized state transformation equation is constructed. The specific optimization calculation process is expressed as follows:
[0072]
[0073]
[0074] Setting both equations to 0, the first and second optimization parameters obtained from the optimization calculation are expressed as follows:
[0075]
[0076]
[0077] Among them, W i B represents the first optimization parameter. i This represents the second optimization parameter. express The mean.
[0078] It should be understood that, although Figure 1The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.
[0079] In one embodiment, such as Figure 2 As shown, a device for predicting the chemical composition of lead-barium glass before weathering is provided, comprising:
[0080] The data sampling module 201 is used to sample and selectively label the chemical composition characteristics of ancient lead-barium glass artifacts to obtain a lead-barium glass dataset; wherein, the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points;
[0081] Data clustering module 202 is used to cluster the special dataset of lead-barium glass according to the K-means clustering algorithm to obtain the weathered and unweathered classifier of lead-barium glass. The ordinary dataset of lead-barium glass is input into the weathered and unweathered classifier of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0082] The prediction module 203 is used to construct a state transformation equation of a multi-dimensional vector mapping based on weathered point data and unweathered point data. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathered point data is input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0083] Specific limitations regarding the device for predicting the pre-weathering chemical composition of lead-barium glass can be found in the limitations of the prediction method for pre-weathering chemical composition of lead-barium glass described above, and will not be repeated here. Each module in the aforementioned device for predicting the pre-weathering chemical composition of lead-barium glass can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.
[0084] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When executed by the processor, the computer program implements a method for predicting the chemical composition of lead-barium glass before weathering. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.
[0085] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.
[0086] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to perform the following steps:
[0087] The chemical composition characteristics of ancient lead-barium glass artifacts were sampled and selectively labeled to obtain a lead-barium glass dataset; the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points.
[0088] The special dataset of lead-barium glass is clustered according to the K-means clustering algorithm to obtain the weathered and unweathered classifiers of lead-barium glass. The ordinary dataset of lead-barium glass is then input into the weathered and unweathered classifiers of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0089] A state transformation equation based on weathered point data and unweathered point data is constructed using a multidimensional vector mapping. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathered point data is then input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0090] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:
[0091] The chemical composition characteristics of ancient lead-barium glass artifacts were sampled and selectively labeled to obtain a lead-barium glass dataset; the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points.
[0092] The special dataset of lead-barium glass is clustered according to the K-means clustering algorithm to obtain the weathered and unweathered classifiers of lead-barium glass. The ordinary dataset of lead-barium glass is then input into the weathered and unweathered classifiers of lead-barium glass for data classification to obtain weathered point data and unweathered point data.
[0093] A state transformation equation based on weathered point data and unweathered point data is constructed using a multidimensional vector mapping. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathered point data is then input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathered point before weathering.
[0094] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0095] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0096] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.
Claims
1. A method for predicting the chemical composition content of lead-barium glass before weathering, characterized in that, The method includes: The chemical composition characteristics of ancient lead-barium glass artifacts were sampled and selectively labeled to obtain a lead-barium glass dataset; wherein, the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points; The special dataset of lead-barium glass is clustered according to the K-means clustering algorithm to obtain a weathered and unweathered classifier for lead-barium glass. The ordinary dataset of lead-barium glass is then input into the weathered and unweathered classifier for data classification to obtain weathered point data and unweathered point data. Based on the weathering point data and the unweathered point data, a state transformation equation of multidimensional vector mapping is constructed. The state transformation equation is optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation. The weathering point data is input into the optimized state transformation equation for prediction to obtain the chemical composition content of the weathering point before weathering corresponding to the weathering point data. A state transformation equation based on the weathered point data and the unweathered point data is constructed using a multi-dimensional vector mapping. The state transformation equation is then optimized by minimizing its mean square error, resulting in an optimized state transformation equation, which includes: Based on weathering point data And unweathered point data Construct the state transformation equation for the multidimensional vector mapping, expressed as follows: ;in, Indicates the first weathering point in the data Content of various chemical components This indicates the first unweathered point in the data. Content of various chemical components and Let these represent the first and second parameters of the state transformation equation, respectively. Indicates the total number of chemical components; The first and second parameters of the state transformation equation are optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation.
2. The method according to claim 1, characterized in that, Chemical composition characteristics of ancient lead-barium glass artifacts were sampled and selectively labeled to obtain a lead-barium glass dataset, including: The chemical composition characteristics of ancient lead-barium glass artifacts were sampled to obtain the types and contents of chemical components at each sampling point; Sampling points are selectively labeled based on their weathering degree to obtain labeled and unlabeled sample points. A special dataset for lead-barium glass is constructed based on the types and contents of chemical components at the labeled sample points, and a general dataset for lead-barium glass is constructed based on the types and contents of chemical components at the unlabeled sample points. The labeled sample points include labeled weathered points and labeled unweathered points.
3. The method according to claim 2, characterized in that, The chemical components at each sampling point include silicon dioxide, sulfur dioxide, phosphorus pentoxide, barium oxide, calcium oxide, potassium oxide, aluminum oxide, magnesium oxide, sodium oxide, lead oxide, strontium oxide, iron oxide, copper oxide, and tin oxide.
4. The method according to claim 1, further comprising, after obtaining the lead-barium glass dataset: Missing chemical composition values in the lead-barium glass dataset are padded with zeros, and the proportion of different chemical compositions at each sampling point in the lead-barium glass dataset is calculated. The cumulative proportions of chemical compositions at the same sampling point and the sampling points with proportions between 85% and 105% are considered as valid sampling points. By statistically analyzing the types and contents of chemical compositions at the valid sampling points, the quality-controlled lead-barium glass dataset is obtained.
5. The method according to claim 1, characterized in that, The specific dataset of lead-barium glass is clustered using the K-means clustering algorithm to obtain classifiers for weathered and unweathered lead-barium glass, including: Two labeled sample points in the lead-barium glass special dataset are randomly selected as centroids. By calculating the distance between each labeled sample point and the two centroids, each labeled sample point is clustered into the cluster to which the nearest centroid belongs. The average value of all labeled sample points in the cluster obtained by clustering is used as the new centroid, and the position of the centroid is updated. Based on the new centroid, the labeled sample points in the lead-barium glass special dataset are repeatedly clustered until the centroid position no longer changes. Then, the clustering stops and the weathered and unweathered classifiers for lead-barium glass are output.
6. The method according to claim 1, characterized in that, The first and second parameters of the state transformation equation are optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation, which includes: The mean square error of the state transformation equation is expressed as: in, This indicates the first unweathered point in the data. The first unweathered point Content of various chemical components Indicating the weathering point data, the first The first weathering point Content of various chemical components m express The total number; By minimizing For the first parameter of the state transformation equation Second parameter Optimization is performed to obtain first and second optimization parameters. Based on these parameters, an optimized state transition equation is constructed, where the first and second optimization parameters are respectively expressed as... in, Indicates the first optimization parameter. This represents the second optimization parameter. express The mean.
7. A device for predicting the chemical composition content of lead-barium glass before weathering, characterized in that, The device includes: The data sampling module is used to sample and selectively label the chemical composition characteristics of ancient lead-barium glass artifacts to obtain a lead-barium glass dataset; wherein, the lead-barium glass dataset includes a special dataset of lead-barium glass containing labeled sample points and a general dataset of lead-barium glass containing unlabeled sample points; The data clustering module is used to cluster the special dataset of lead-barium glass according to the K-means clustering algorithm to obtain a weathered and unweathered classifier for lead-barium glass. The ordinary dataset of lead-barium glass is input into the weathered and unweathered classifier for data classification to obtain weathered point data and unweathered point data. The prediction module is used to construct a state transformation equation of multidimensional vector mapping based on the weathering point data and the unweathered point data, optimize the state transformation equation by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation, input the weathering point data into the optimized state transformation equation for prediction, and obtain the chemical composition content of the weathering point before weathering corresponding to the weathering point data. A state transformation equation based on the weathered point data and the unweathered point data is constructed using a multi-dimensional vector mapping. The state transformation equation is then optimized by minimizing its mean square error, resulting in an optimized state transformation equation, which includes: Based on weathering point data And unweathered point data Construct the state transformation equation for the multidimensional vector mapping, expressed as follows: ;in, Indicates the first weathering point in the data Content of various chemical components This indicates the first unweathered point in the data. Content of various chemical components and Let these represent the first and second parameters of the state transformation equation, respectively. Indicates the total number of chemical components; The first and second parameters of the state transformation equation are optimized by minimizing the mean square error of the state transformation equation to obtain the optimized state transformation equation.
8. 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 6.
9. 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 6.