A method and system for analyzing mechanical wear particle materials and wear parts

Abstract

The present invention discloses a method for analyzing mechanical wear particle materials and wear parts in the field of data analysis technology, comprising the following steps: 1) sampling lubricating oil; 2) analyzing the oil sample using an energy spectrum analyzer to obtain energy spectrum analysis data; 3) preprocessing the energy spectrum analysis data; 4) establishing a material component library, which includes all materials involved in lubricating parts of mechanical equipment; 5) performing particle material analysis calculations on the energy spectrum analysis data and the material component library to obtain material analysis results; 6) establishing a parts material library, which includes mechanical lubricating parts and their corresponding material grades; 7) querying the parts material library according to the material grade to obtain the wear parts; 8) outputting the material grade, material name, wear parts and confidence level in descending order of confidence. The present invention improves the accuracy and automation level of mechanical wear particle material and wear part analysis based on energy spectrum data.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to a mechanical wear particle material and a wear part analysis method. Background Art

[0002] When oil-lubricated parts of mechanical equipment wear out, wear particles will appear in the lubricating oil. The chemical composition and content of the wear particles can be measured by an energy spectrometer. Further analysis of the chemical composition and content can reveal the material of the wear particles. Based on the correspondence between parts and materials, the wear location can be inferred, allowing for condition-based maintenance.

[0003] Currently, wear particle material and wear site analysis is primarily performed manually, which suffers from heavy reliance on expert experience and low analysis efficiency. Automating this analysis presents three practical challenges.

[0004] (1) Oxide treatment issues. Particles in lubricating oil are easily oxidized. Oxidized particles usually have a high oxygen content. If they are directly matched with the chemical composition of the material, it is easy to cause material inference errors.

[0005] (2) Inaccurate chemical composition content measured by the energy spectrometer. The chemical composition content of the particles obtained by the energy spectrometer is not very accurate. If it is directly matched with the chemical composition content range of the material, it is easy to conclude that the wear particles do not belong to any material.

[0006] (3) The confidence level of the inferred conclusion. When analyzing the material of a particle, it is often impossible to confirm the material to which the particle belongs 100%. In many cases, only a range of the material to which the particle belongs can be given. Therefore, if the confidence level of the inferred conclusion can be given, it will be very helpful for practical applications. Summary of the Invention

[0007] In response to the deficiencies in the prior art, the present invention provides a method for analyzing mechanical wear particle materials and wear parts, which solves the problems of oxide processing, inaccurate chemical composition content measured by energy spectrometer, and confidence level of inferred conclusions.

[0008] The object of the present invention is achieved as follows: A mechanical wear particulate material and wear site analysis method is characterized by comprising the following steps:

[0009] Step 1) regularly sampling lubricating oil from the lubrication system of the mechanical equipment;

[0010] Step 2) Analyzing the oil sample using an energy spectrum analyzer to obtain energy spectrum analysis data;

[0011] Step 3) preprocessing the energy spectrum analysis data in step 2) to obtain preprocessed energy spectrum analysis data;

[0012] Step 4) Establish a material composition library, which includes all materials involved in the lubrication parts of mechanical equipment. For each material, it should include material brand, material name, all chemical components and their lower and upper limits;

[0013] Step 5) performing particle material analysis calculation on the energy spectrum analysis data preprocessed in step 3) and the material composition library in step 4) to obtain material analysis results, including material brand, material name and confidence level;

[0014] Step 6) Establishing a parts material library, which includes mechanical lubrication parts and their corresponding material grades;

[0015] Step 7) According to the material brand, query the parts material library to obtain the wear parts;

[0016] Step 8) Output the material brand, material name, wear location and confidence level in descending order of confidence level.

[0017] As a preferred technical solution for the mechanical wear particle material and wear part analysis method described in the present invention, the pretreatment method in step 3) specifically includes: first determining the oxygen content; if it is less than 10%, no treatment is performed; if it is greater than or equal to 10%, first marking the particle as an oxide, then setting the oxygen content to zero, renormalizing the chemical component content, and using the normalized energy spectrum analysis data as the preprocessed energy spectrum analysis data.

[0018] As a preferred technical solution of the mechanical wear particulate material and wear site analysis method of the present invention, the method for analyzing and calculating the particulate material in step 5) is specifically as follows:

[0019] Step 5-1) Assume that the number of material types in the material composition library is n, the number of chemical composition types of all materials is k, the lower limit of the content of the jth chemical component of the i-th material is lower[i][j], the upper limit of the content is upper[i][j], the content of the jth chemical component in the energy spectrum analysis data is data[j], the matching score of the i-th material is score[i], and the proportional factor for the expansion of the material chemical composition range is Δx;

[0020] Step 5-2) Initialize the scaling factor Δx for material chemical composition range expansion to 0;

[0021] Step 5-3) Initialize the matching scores of all materials to 0, that is, let score[i] = 0, i = 1, 2, ..., n;

[0022] Step 5-4) Take the i-th material from the material library;

[0023] Step 5-5) According to formula (1) and formula (2), the content range of the jth chemical component of the i-th material is expanded from [lower[i][j], upper[i][j]] to [lower*[i][j], upper*[i][j]];

[0024] lower*[i][j]=max(lower[i][j]-lower[i][j]×Δx,0) (1)

[0025] upper*[i][j]=min(upper[i][j]+upper[i][j]×Δx,100) (2)

[0026] Step 5-6) From the energy spectrum analysis data, obtain the content data[j] of the jth chemical component;

[0027] Step 5-7) According to formula (3), determine whether the content of the j-th chemical component measured by the energy spectrum analysis falls within the expanded interval. If so, according to formula (4), add the content of the j-th chemical component to the matching score of the i-th material; otherwise, repeat steps 5-5) to 5-7) to match the next chemical component. Step 5-5) to 5-7) are repeated k times to complete the matching of all chemical components, thereby obtaining the matching score of the i-th material. The matching score of each material is used as the confidence level for inferring the material, and then the matching score of the next material is calculated. Step 5-4) to 5-7) are repeated n times to obtain the matching scores of all materials in the material library.

[0028] lower*[i][j]≤data[j]≤upper*[i][j] (3)

[0029] score[i]=score[i]+data[j] (4)

[0030] Step 5-8) Calculate the sum of the matching scores of all materials and compare the sum with the pre-set matching stop threshold T. If the sum of the matching scores is not less than the stop threshold T, output the matching scores of all materials and the proportional factor Δx for chemical composition range expansion, and the particle material analysis is completed. Otherwise, according to formula (5), increase the proportional factor Δx for chemical composition range expansion of the material, step = 0.01, and then repeat steps 5-3) to 5-8) to recalculate the matching scores of all materials until the sum of the matching scores of all materials is greater than or equal to the matching stop threshold T.

[0031] Δx=Δx+step (5).

[0032] A mechanical wear particle material and wear site analysis system, used to perform the above method, comprising:

[0033] Database module, used to store relevant information of mechanical equipment, oil sample information, energy spectrum analysis data, chemical composition information of materials, analysis results of wear particle materials and wear parts;

[0034] User registration module, used to register users who can use the system;

[0035] Equipment registration module, used to register equipment model, equipment number, parts, materials and other equipment information required by the system;

[0036] The oil sample registration module is used to generate an oil sample number based on the relevant information of the mechanical equipment and the relevant information of the oil sample, and to create a record for the oil sample in the database;

[0037] Data import module, used to import spectrum analysis data from WORD document into database;

[0038] The data query module is used to query the energy spectrum analysis data imported into the database according to the specified query conditions;

[0039] Data analysis module, used to analyze the material and wear location of wear particles based on energy spectrum analysis data;

[0040] The report generation module is used to generate an analysis report on the mechanical wear particle material and wear location of the current oil sample.

[0041] As a preferred technical solution of the mechanical wear particle material and wear part analysis system described in the present invention, before using the system, user registration and device registration must be completed first;

[0042] User registration is to determine who can use the system, including user name and password;

[0043] Equipment registration is to register the object of mechanical wear monitoring, including equipment model, equipment number, parts information, material information and lubricant information, etc.

[0044] After completing the registration of the above two basic information, for an oil sample to be analyzed, first generate an oil sample number based on the relevant information of the mechanical equipment and the relevant information of the oil sample, and create a record for the oil sample in the database to complete the oil sample registration; then import the energy spectrum analysis data corresponding to the oil sample into the database. If mechanical wear particle material and wear part analysis is to be performed, first query the energy spectrum analysis data corresponding to the oil sample according to the relevant conditions. After querying the relevant data, the corresponding analysis can be performed. After the analysis is completed, the analysis results can be saved to the database; if an oil sample has been analyzed, an analysis report in WORD format can be generated.

[0045] Compared with the prior art, the present invention has the beneficial effect of improving the accuracy and automation level of the analysis of mechanical wear particle materials and wear parts based on energy spectrum data. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0047] Figure 1 Flow chart of the analysis method of the present invention.

[0048] Figure 2 This is a flow chart of the pretreatment process in the analysis method of the present invention.

[0049] Figure 3 This is a flow chart for analyzing particulate materials in the analytical method of the present invention.

[0050] Figure 4 This is a module diagram of the analysis system in the present invention.

[0051] Figure 5 This is the functional logic diagram of the analysis system in the present invention.

[0052] Figure 6 This is a screenshot of the analysis system software interface in the present invention. Figure 1 .

[0053] Figure 7 This is a screenshot of the analysis system software interface in the present invention. Figure 2 . DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] like Figure 1-3 A method for analyzing mechanical wear particle materials and wear sites is shown, comprising the following steps.

[0056] Step 1) According to the wear monitoring plan of the mechanical equipment, lubricating oil samples are regularly taken from the lubrication system of the mechanical equipment.

[0057] Step 2) Analyze the oil sample using an energy spectrum analyzer to obtain energy spectrum analysis data.

[0058] Step 3) preprocessing the energy spectrum analysis data in step 2) to obtain preprocessed energy spectrum analysis data; the preprocessing method specifically includes: first determining the oxygen content; if it is less than 10%, no processing is performed; if it is greater than or equal to 10%, first marking the particle as an oxide, then setting the oxygen content to zero, re-normalizing the chemical component content, and using the normalized energy spectrum analysis data as the preprocessed energy spectrum analysis data.

[0059] Step 4) Establish a material composition library, which includes all materials involved in the lubrication parts of mechanical equipment. For each material, it should include material brand, material name, all chemical components and their lower and upper limits.

[0060] Step 5) performing particle material analysis calculation on the energy spectrum analysis data preprocessed in step 3) and the material composition library in step 4) to obtain material analysis results, including material brand, material name and confidence level; the particle material analysis calculation method is specifically as follows:

[0061] Step 5-1) Assume that the number of material types in the material composition library is n, the number of chemical composition types of all materials is k, the lower limit of the content of the jth chemical component of the i-th material is lower[i][j], the upper limit of the content is upper[i][j], the content of the jth chemical component in the energy spectrum analysis data is data[j], the matching score of the i-th material is score[i], and the proportional factor for the expansion of the material chemical composition range is Δx;

[0062] Step 5-2) Initialize the scaling factor Δx for material chemical composition range expansion to 0;

[0063] Step 5-3) Initialize the matching scores of all materials to 0, that is, let score[i] = 0, i = 1, 2, ..., n;

[0064] Step 5-4) Take the i-th material from the material library;

[0065] Step 5-5) According to formula (1) and formula (2), the content range of the jth chemical component of the i-th material is expanded from [lower[i][j], upper[i][j]] to [lower*[i][j], upper*[i][j]];

[0066] lower*[i][j]=max(lower[i][j]-lower[i][j]×Δx,0) (1)

[0067] upper*[i][j]=min(upper[i][j]+upper[i][j]×Δx,100) (2)

[0068] Step 5-6) From the energy spectrum analysis data, obtain the content data[j] of the jth chemical component;

[0069] Step 5-7) According to formula (3), determine whether the content of the j-th chemical component measured by the energy spectrum analysis falls within the expanded interval. If so, according to formula (4), add the content of the j-th chemical component to the matching score of the i-th material; otherwise, repeat steps 5-5) to 5-7) to match the next chemical component. Step 5-5) to 5-7) are repeated k times to complete the matching of all chemical components, thereby obtaining the matching score of the i-th material. The matching score of each material is used as the confidence level for inferring the material, and then the matching score of the next material is calculated. Step 5-4) to 5-7) are repeated n times to obtain the matching scores of all materials in the material library.

[0070] lower*[i][j]≤data[j]≤upper*[i][j] (3)

[0071] score[i]=score[i]+data[j] (4)

[0072] Step 5-8) Calculate the sum of the matching scores of all materials and compare the sum result with the pre-set matching stop threshold T. If the sum of the matching scores is not less than the stop threshold T (based on practical experience, T can be 100), then output the matching scores of all materials and the proportional factor Δx for chemical composition range expansion, and the particle material analysis is completed; otherwise, according to formula (5), increase the proportional factor Δx for chemical composition range expansion of the material, step = 0.01, and then repeat steps 5-3) to 5-8) to recalculate the matching scores of all materials until the sum of the matching scores of all materials is greater than or equal to the matching stop threshold T;

[0073] Δx=Δx+step (5).

[0074] Step 6) Establish a parts material library, which includes mechanical lubrication parts and their corresponding material grades.

[0075] Step 7) According to the material brand, query the parts material library to obtain the wear parts.

[0076] Step 8) Output the material brand, material name, wear location and confidence level in descending order of confidence level.

[0077] A mechanical wear particle material and wear site analysis system, used to perform the above method, comprising:

[0078] Database module, used to store relevant information of mechanical equipment, oil sample information, energy spectrum analysis data, chemical composition information of materials, analysis results of wear particle materials and wear parts;

[0079] User registration module, used to register users who can use the system;

[0080] Equipment registration module, used to register equipment model, equipment number, parts, materials and other equipment information required by the system;

[0081] The oil sample registration module is used to generate an oil sample number based on the relevant information of the mechanical equipment and the relevant information of the oil sample, and to create a record for the oil sample in the database;

[0082] Data import module, used to import spectrum analysis data from WORD document into database;

[0083] The data query module is used to query the energy spectrum analysis data imported into the database according to the specified query conditions;

[0084] Data analysis module, used to analyze the material and wear location of wear particles based on energy spectrum analysis data;

[0085] The report generation module is used to generate an analysis report on the mechanical wear particle material and wear location of the current oil sample.

[0086] Before using the system, you first need to complete user registration and device registration;

[0087] User registration is to determine who can use the system, including user name and password;

[0088] Equipment registration is to register the object of mechanical wear monitoring, including equipment model, equipment number, parts information, material information and lubricant information, etc.

[0089] After completing the registration of the above two basic information, for an oil sample to be analyzed, first generate an oil sample number based on the relevant information of the mechanical equipment and the relevant information of the oil sample, and create a record for the oil sample in the database to complete the oil sample registration; then import the energy spectrum analysis data corresponding to the oil sample into the database. If mechanical wear particle material and wear part analysis is to be performed, first query the energy spectrum analysis data corresponding to the oil sample according to the relevant conditions. After querying the relevant data, the corresponding analysis can be performed. After the analysis is completed, the analysis results can be saved to the database; if an oil sample has been analyzed, an analysis report in WORD format can be generated.

[0090] The present invention is further described below with reference to a specific example. This example uses the present invention to analyze the primary wear particle material and wear parts of a certain mechanical equipment. The specific implementation process is as follows:

[0091] Step 1) Register the relevant information of the mechanical equipment in the system, mainly including equipment model, equipment number, parts information, material information and lubricant information;

[0092] Step 2) sampling lubricating oil from the lubrication system of the mechanical equipment, generating an oil sample number for the oil sample in the system, registering the oil sample information, and creating a record for the oil sample in the system database after registration is completed;

[0093] Step 3) Analyze the oil sample using a scanning electron microscope / energy spectrum analyzer to obtain a scanning electron microscope image, energy spectrum image, and energy spectrum data, and then import the relevant data into the system database;

[0094] Step 4) When analyzing the data, first query the data based on the device model, device number, sampling location and sampling time interval, such as Figure 6 Click the "Particle Intelligent Analysis" button in the upper right corner to enter the following Figure 7 The analysis interface shown;

[0095] Step 5) Click the "Material Analysis" button to complete the analysis of the data. The analysis results are as follows: Figure 7 As shown in the lower part, including material brand, material name, material matching score (confidence) and wear location, the results are listed from high to low according to the material matching score. The highest matching score is 93.88, including two materials M50 and 8Cr4Mo4V, both of which are bearing steels; the professional analysis result of the data is M50, and the analysis result of the present invention is consistent with the analysis of the professionals.

[0096] The above embodiments are only intended to help understand the method and core concept of the present invention. It should be noted that, without departing from the principles of the present invention, a number of improvements and modifications may be made to the present invention by those skilled in the art, and such improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A method for analyzing mechanical wear particle materials and wear sites, characterized in that: The following steps are involved: Step 1) regularly sampling lubricating oil from the lubrication system of the mechanical equipment; Step 2) Analyzing the oil sample using an energy spectrum analyzer to obtain energy spectrum analysis data; Step 3) preprocessing the energy spectrum analysis data in step 2) to obtain preprocessed energy spectrum analysis data; Step 4) Establish a material composition library, which includes all materials involved in the lubrication parts of mechanical equipment. For each material, it should include material brand, material name, all chemical components and their lower and upper limits; Step 5) The energy spectrum analysis data pre-processed in step 3) and the material composition library in step 4) are used to perform particle material analysis calculations to obtain material analysis results, including material brand, material name, and confidence level. The particle material analysis calculation method is specifically as follows: Step 5-1) Assume that the number of material types in the material composition library is n, the number of chemical composition types of all materials is k, the lower limit of the content of the jth chemical component of the i-th material is lower[i][j], the upper limit of the content is upper[i][j], the content of the jth chemical component in the energy spectrum analysis data is data[j], the matching score of the i-th material is score[i], and the proportional factor for the expansion of the material chemical composition range is Δx; Step 5-2) Initialize the scaling factor Δx for material chemical composition range expansion to 0; Step 5-3) Initialize the matching scores of all materials to 0, that is, let score[i] = 0, i = 1, 2, ..., n; Step 5-4) Take the i-th material from the material library; Step 5-5) According to formula (1) and formula (2), the content range of the jth chemical component of the i-th material is expanded from [lower[i][j], upper[i][j]] to [lower*[i][j], upper*[i][j]]; lower*[i][j]=max(lower[i][j]-lower[i][j]×Δx,0) (1) upper*[i][j]=min(upper[i][j]+upper[i][j]×Δx,100) (2) Step 5-6) From the energy spectrum analysis data, obtain the content data[j] of the jth chemical component; Step 5-7) According to formula (3), determine whether the content of the j-th chemical component measured by the energy spectrum analysis falls within the expanded interval. If so, according to formula (4), add the content of the j-th chemical component to the matching score of the i-th material; otherwise, repeat steps 5-5) to 5-7) to match the next chemical component. Step 5-5) to 5-7) are repeated k times to complete the matching of all chemical components, thereby obtaining the matching score of the i-th material. The matching score of each material is used as the confidence level for inferring the material, and then the matching score of the next material is calculated. Step 5-4) to 5-7) are repeated n times to obtain the matching scores of all materials in the material library. lower*[i][j]≤data[j]≤upper*[i][j] (3) score[i]=score[i]+data[j] (4) Step 5-8) Calculate the sum of the matching scores of all materials and compare the sum result with the pre-set matching stop threshold T. If the sum of the matching scores is not less than the stop threshold T, then output the matching scores of all materials and the proportional factor Δx for chemical composition range expansion, and the particle material analysis is completed; otherwise, according to formula (5), increase the proportional factor Δx for chemical composition range expansion of the material, step = 0.01, and then repeat steps 5-3) to 5-8) to recalculate the matching scores of all materials until the sum of the matching scores of all materials is greater than or equal to the matching stop threshold T; Δx=Δx+step (5); Step 6) Establishing a parts material library, which includes mechanical lubrication parts and their corresponding material grades; Step 7) According to the material brand, query the parts material library to obtain the wear parts; Step 8) Output the material brand, material name, wear location and confidence level in descending order of confidence level.

2. A mechanical wear particle material and wear site analysis method according to claim 1, characterized in that: The pretreatment method in step 3) specifically includes: first determining the oxygen content; if it is less than 10%, no treatment is performed; if it is greater than or equal to 10%, the particle is first marked as an oxide, and then the oxygen content is set to zero, and the chemical composition content is renormalized, and the normalized energy spectrum analysis data is used as the preprocessed energy spectrum analysis data.

3. A mechanical wear particle material and wear part analysis system, characterized in that: It is used to perform the method according to any one of claims 1 to 2, comprising: Database module, used to store relevant information of mechanical equipment, oil sample information, energy spectrum analysis data, chemical composition information of materials, analysis results of wear particle materials and wear parts; User registration module, used to register users who can use the system; Equipment registration module, used to register equipment model, equipment number, parts, and equipment information required by the material system; The oil sample registration module is used to generate an oil sample number based on the relevant information of the mechanical equipment and the relevant information of the oil sample, and to create a record for the oil sample in the database; Data import module, used to import spectrum analysis data from WORD document into database; The data query module is used to query the energy spectrum analysis data imported into the database according to the specified query conditions; Data analysis module, used to analyze the material and wear location of wear particles based on energy spectrum analysis data; The report generation module is used to generate an analysis report on the mechanical wear particle material and wear location of the current oil sample.

4. The mechanical wear particle material and wear part analysis system according to claim 3, characterized in that: Before using the system, you first need to complete user registration and device registration; User registration determines who can use the system, including username and password; Equipment registration is to register the object of mechanical wear monitoring, including equipment model, equipment number, parts information, material information and lubricant information; After completing the registration of the above two basic information, for each oil sample to be analyzed, first generate an oil sample number based on the relevant information of the mechanical equipment and the relevant information of the oil sample, and create a record for the oil sample in the database to complete the oil sample registration; Then import the energy spectrum analysis data corresponding to the oil sample into the database. If you want to analyze the mechanical wear particle materials and wear parts, you first need to query the energy spectrum analysis data corresponding to the oil sample according to relevant conditions. After querying the relevant data, you can perform the corresponding analysis. After the analysis is completed, you can save the analysis results to the database; if an oil sample has been analyzed, you can generate an analysis report in WORD format.

Images