Material type identification system and method applied to automatic operation of sand mill

Abstract

The invention provides a material type identification system and method applied to automatic operation of a sand mill, and the system comprises a material type identification subsystem which is used for obtaining a material type identification result of a to-be-ground material on an automatic operation assembly line of the sand mill; the grinding strategy calling subsystem is used for calling a grinding strategy library according to the material type identification result and determining a grinding strategy for identifying the material; and the automatic grinding subsystem is used for controlling a sand mill to automatically grind the identified materials based on the grinding strategy. According to the material type identification system and method applied to automatic operation of the sand mill, the material type identification result of the to-be-ground material is obtained, the grinding strategy library is called according to the material type identification result, the grinding strategy for identifying the material is determined, and the sand mill is automatically controlled to grind based on the grinding strategy; corresponding conveying of materials to different types of sand mills and cooperative grinding control are achieved, and the working efficiency of the sand mills is greatly improved.

Description

Technical Field

[0001] The invention relates to the technical field of sand mills, and in particular to a material type identification system and method applied to the automated operation of sand mills. Background Art

[0002] A sand mill is a high-efficiency wet ultrafine grinding and dispersing equipment. During the grinding process, different types of materials have different physical and chemical properties. Traditional sand mills often rely on manual adjustment of control parameters according to the type of material (for example, materials with higher hardness require greater grinding force and longer time, while softer materials can be processed quickly). Manual operation has the risk of misoperation, and the operator's workload is heavy and the work efficiency is low.

[0003] In view of this, there is an urgent need for a material type identification system and method for sand mill automation operation to at least solve the above-mentioned shortcomings. Summary of the invention

[0004] One of the purposes of the present invention is to provide a material type identification system and method for the automated operation of a sand mill, to obtain the material type identification result of the material to be ground, to retrieve the grinding strategy library according to the material type identification result, to determine the grinding strategy of the identified material, and to automatically control the sand mill grinding based on the grinding strategy, thereby realizing the corresponding delivery and coordinated grinding control of the identified material to different types of sand mills, and greatly improving the working efficiency of the sand mill.

[0005] An embodiment of the present invention provides a material type identification system for sand mill automation operation, comprising:

[0006] The material type identification subsystem is used to obtain the material type identification result of the material to be ground on the sand mill automated production line;

[0007] The grinding strategy retrieval subsystem is used to retrieve the grinding strategy library according to the material type identification result and determine the grinding strategy of the identified material;

[0008] The automatic grinding subsystem is used to control the sand mill to automatically grind the identified materials based on the grinding strategy.

[0009] Preferably, the material type identification subsystem obtains the material type identification result of the material to be ground on the sand mill automated production line and performs the following operations:

[0010] Based on machine vision technology, the image of the material to be ground is obtained and the image features are extracted. According to the image features, the material type recognition result is determined;

[0011] and / or,

[0012] Based on spectral analysis technology, the spectral characteristic information of the material to be ground is analyzed, and then the material type identification result is determined according to the spectral characteristic information;

[0013] and / or,

[0014] The material type recognition model is trained based on the machine learning algorithm and combined with historical data, and then the material type recognition result is determined based on the real-time sensor data and the material type recognition model.

[0015] Preferably, the grinding strategy retrieval subsystem retrieves the grinding strategy library according to the material type identification result, determines the grinding strategy of the identified material, and performs the following operations:

[0016] According to the material type identification result, determine the type of sand mill suitable for grinding the identified material;

[0017] Get the grinding strategy library of the sand mill of the sand mill type;

[0018] Characterize the material type recognition result and obtain the recognition result feature value set;

[0019] The grinding strategy is determined according to the feature value set of the recognition result and the grinding strategy library.

[0020] Preferably, the grinding strategy calling subsystem obtains the grinding strategy library of the sand mill type and performs the following operations:

[0021] Obtain the control records of the operators of the sand mills of the sand mill type;

[0022] Analyze the operation record, obtain the manual recognition result and grinding parameters of the manipulated material, characterize the manual recognition result, obtain a first manipulated material recognition result feature value set, and associate it with the grinding parameters to obtain a first associated item;

[0023] Perform machine learning on the manual recognition results and grinding parameters of the same controlled material to obtain a grinding parameter determination model;

[0024] Determine missing recognition results based on manual recognition results;

[0025] Inputting the missing identification result into the grinding parameter determination model, obtaining the missing grinding parameter output by the grinding parameter determination model, characterizing the missing identification result, obtaining a second manipulated material identification result feature value set, and associating it with the missing grinding parameter to obtain a second associated item;

[0026] The first associated items and the second associated items of all the manipulated materials are integrated to obtain a grinding strategy library of the sand mill of the sand mill type.

[0027] The material type identification system for sand mill automation operation provided by the embodiment of the present invention also includes:

[0028] The verification subsystem is used to obtain the sampled material after the identified material is ground, and to verify the grinding accuracy of the sampled material. If the grinding accuracy verification fails, the grinding medium is tested.

[0029] Preferably, when the grinding accuracy verification fails, the verification subsystem performs a grinding medium test and performs the following operations:

[0030] Obtain real-time triboelectric signals between friction pairs;

[0031] Obtaining initial triboelectric signals of the friction pair;

[0032] Calculating the change in friction coefficient according to the electric signal difference between the real-time triboelectric signal and the initial triboelectric signal;

[0033] If the change in the friction coefficient reaches a preset change degree, the corresponding grinding medium is used as the determination grinding medium, and the use time of the determination grinding medium is obtained;

[0034] Determine the rationality of the use time of the grinding media. If it is unreasonable, reset the cleaning time setting parameters;

[0035] Among them, the rationality of the use time of the grinding medium is judged. If it is unreasonable, the cleaning time setting parameters are reset, including:

[0036] Obtain the standard usage time for determining the grinding media;

[0037] If the time difference between the standard usage time and the usage time is greater than or equal to the preset time difference threshold, the cleaning time setting parameter is obtained;

[0038] Obtain a comparison table of cleaning time setting parameter downward adjustment values;

[0039] According to the comparison table of the time difference and the cleaning time setting parameter downward adjustment value, determine the target cleaning time setting parameter after downward adjustment;

[0040] After notifying the staff to replace the determined grinding media, the target cleaning time setting parameters should be applied.

[0041] An embodiment of the present invention provides a material type identification method applied to the automated operation of a sand mill, comprising:

[0042] Step 1: Obtain the material type identification result of the material to be ground on the sand mill automation production line;

[0043] Step 2: Retrieve the grinding strategy library according to the material type identification result and determine the grinding strategy for the identified material;

[0044] Step 3: Based on the grinding strategy, control the sand mill to automatically grind the identified material.

[0045] Preferably, obtaining the material type identification result of the material to be ground on the sand mill automated production line includes:

[0046] Based on machine vision technology, the image of the material to be ground is obtained and the image features are extracted. According to the image features, the material type recognition result is determined;

[0047] and / or,

[0048] Based on spectral analysis technology, the spectral characteristic information of the material to be ground is analyzed, and then the material type identification result is determined according to the spectral characteristic information;

[0049] and / or,

[0050] The material type recognition model is trained based on the machine learning algorithm and combined with historical data, and then the material type recognition result is determined based on the real-time sensor data and the material type recognition model.

[0051] Preferably, the grinding strategy library is retrieved according to the material type identification result to determine the grinding strategy of the identified material, including:

[0052] According to the material type identification result, determine the type of sand mill suitable for grinding the identified material;

[0053] Get the grinding strategy library of the sand mill of the sand mill type;

[0054] Characterize the material type recognition result and obtain the recognition result feature value set;

[0055] The grinding strategy is determined according to the feature value set of the recognition result and the grinding strategy library.

[0056] The embodiment of the present invention provides a material type identification method applied to the automatic operation of a sand mill, further comprising:

[0057] Obtain sampled material after grinding the identified material, perform grinding accuracy verification on the sampled material, and perform grinding medium detection when the grinding accuracy verification fails.

[0058] Preferably, when the grinding accuracy check fails, a grinding medium test is performed, including:

[0059] Obtain real-time triboelectric signals between friction pairs;

[0060] Obtaining initial triboelectric signals of the friction pair;

[0061] Calculating the change in friction coefficient according to the electric signal difference between the real-time triboelectric signal and the initial triboelectric signal;

[0062] If the change in the friction coefficient reaches a preset change degree, the corresponding grinding medium is used as the determination grinding medium, and the use time of the determination grinding medium is obtained;

[0063] Determine the rationality of the use time of the grinding media. If it is unreasonable, reset the cleaning time setting parameters;

[0064] Among them, the rationality of the use time of the grinding medium is judged. If it is unreasonable, the cleaning time setting parameters are reset, including:

[0065] Obtain the standard usage time for determining the grinding media;

[0066] If the time difference between the standard usage time and the usage time is greater than or equal to the preset time difference threshold, the cleaning time setting parameter is obtained;

[0067] Obtain a comparison table of cleaning time setting parameter downward adjustment values;

[0068] According to the comparison table of the time difference and the cleaning time setting parameter downward adjustment value, determine the target cleaning time setting parameter after downward adjustment;

[0069] After notifying the staff to change the grinding media, the target cleaning time setting parameters should be applied.

[0070] The beneficial effects of the present invention are:

[0071] The present invention obtains a material type identification result of a material to be ground, retrieves a grinding strategy library according to the material type identification result, determines a grinding strategy for the identified material, and automatically controls the sand mill grinding based on the grinding strategy, thereby realizing corresponding transportation and coordinated grinding control of the identified material to different types of sand mills, and greatly improving the working efficiency of the sand mill.

[0072] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.

[0073] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0075] Figure 1 It is a schematic diagram of a material type identification system applied to the automated operation of a sand mill according to an embodiment of the present invention;

[0076] Figure 2 It is a schematic diagram of a material type identification method applied to the automated operation of a sand mill in an embodiment of the present invention. DETAILED DESCRIPTION

[0077] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0078] The embodiment of the present invention provides a material type identification system for sand mill automation operation, such as Figure 1 As shown, including:

[0079] The material type identification subsystem 1 is used to obtain the material type identification result of the material to be ground on the sand mill automatic operation line; wherein the sand mill automatic operation line includes sand mills of various types, such as: basket sand mill, horizontal disc sand mill and ultra-fine pin sand mill, etc.; the material to be ground is: the material to be ground, such as: coating, ink and pigment, etc.; the material type identification result is: the material type and material characteristic information (such as: chemical composition, physical characteristics (such as particle size, density, viscosity, etc.), color, etc.) of the material to be ground;

[0080] The grinding strategy retrieval subsystem 2 is used to retrieve the grinding strategy library according to the material type identification result, and determine the grinding strategy of the identified material; wherein the retrieved grinding strategy library is: a database storing grinding control parameters of various material types, each material type has a corresponding grinding strategy, such as: the type of grinder used, the control parameters of the grinder (such as: grinding force and grinding time), etc. The grinding strategy library is constructed based on experience by staff who have experience in local grinder control;

[0081] The automatic grinding subsystem 3 is used to control the sand mill to automatically grind the identified materials based on the grinding strategy.

[0082] The working principle and beneficial effects of the above technical solution are:

[0083] The present invention obtains a material type identification result of a material to be ground, retrieves a grinding strategy library according to the material type identification result, determines a grinding strategy for the identified material, and automatically controls the sand mill grinding based on the grinding strategy, thereby realizing corresponding transportation and coordinated grinding control of the identified material to different types of sand mills, and greatly improving the working efficiency of the sand mill.

[0084] In one embodiment, the material type identification subsystem obtains the material type identification result of the material to be ground on the sand mill automated production line and performs the following operations:

[0085] Based on machine vision technology, the captured image of the material to be ground is obtained and the image features are extracted. According to the image features, the material type recognition result is determined; wherein the image features are: the color, texture, shape and edge of the material to be ground;

[0086] and / or,

[0087] Based on the spectral analysis technology, the spectral characteristic information of the material to be ground is analyzed, and then the material type identification result is determined according to the spectral characteristic information; wherein the spectral characteristic information is: the spectral absorption or emission characteristics of the material to be ground within a specific wavelength range;

[0088] and / or,

[0089] Based on the machine learning algorithm and combined with historical data, the material type recognition model is trained, and then the material type recognition result is determined based on the real-time sensor data and the material type recognition model. When the material type recognition model is trained in combination with historical data, the historical data includes: the sensor historical sensing data and the corresponding historical material type recognition results. The sensor historical sensing data is used as the input of machine learning, and the historical material type recognition results are used as the output of machine learning. After learning all the historical data, the learned material type recognition model is obtained, and the real-time sensor data is input into the material type recognition model to obtain the material type recognition result.

[0090] The working principle and beneficial effects of the above technical solution are:

[0091] The present invention introduces three methods, namely machine vision technology, spectral analysis technology and machine learning, to determine the material type recognition results, thereby improving the comprehensiveness and recognition accuracy of material type recognition.

[0092] In one embodiment, the grinding strategy retrieval subsystem retrieves the grinding strategy library according to the material type identification result, determines the grinding strategy of the identified material, and performs the following operations:

[0093] According to the material type identification result, determine the type of sand mill suitable for grinding the identified material;

[0094] Acquire a grinding strategy library of a sand mill of the sand mill type; wherein the grinding strategy library stores a one-to-one correspondence of a pre-selected recognition result feature value set and a control parameter of the sand mill of the sand mill type;

[0095] Characterize the material type recognition result and obtain the recognition result feature value set;

[0096] The grinding strategy is determined according to the feature value set of the recognition result and the grinding strategy library.

[0097] The working principle and beneficial effects of the above technical solution are:

[0098] Different types of sand mills are suitable for grinding different types of materials (for example, basket sand mills are suitable for grinding medium and low viscosity materials, and horizontal sand mills are suitable for products with high viscosity and fine particle size requirements). Therefore, first, the type of sand mill to be used is roughly determined based on the material type identification result, and then the specific sand mill control parameters are determined to improve the efficiency of obtaining subsequent grinding strategies. After determining the type of sand mill, the grinding strategy library corresponding to the type of sand mill is retrieved. The grinding strategy library stores a one-to-one corresponding pre-selected identification result feature value set and the control parameters of the sand mill of the type of sand mill. The identification result feature value set obtained by characterizing the material type identification result and the pre-selected identification result feature value set are feature matched. When the match is satisfactory, the control parameters of the corresponding sand mill are the grinding strategy.

[0099] In one embodiment, the grinding strategy calling subsystem obtains the grinding strategy library of the sand mill type and performs the following operations:

[0100] Obtaining the control record of the operator of the sand mill of the sand mill type; wherein the control record is: the record generated by the sand mill operator during the actual operation process, including: material information (the operator's identification result of the controlled material, including the material type, chemical and physical state) and grinding parameter settings, and the control record can be an automatic record of the sand mill control system or a manual record;

[0101] Analyze the operation record, obtain the manual recognition result and grinding parameters of the manipulated material, characterize the manual recognition result, obtain a first manipulated material recognition result feature value set, and associate it with the grinding parameters to obtain a first associated item; wherein the process of characterizing the manual recognition result is the same as the process of characterizing the material type recognition result;

[0102] Performing machine learning on the manual recognition results and grinding parameters of the same manipulated material to obtain a grinding parameter determination model; wherein when training the grinding parameter determination model, the input is the manual recognition result and the output is the grinding parameter;

[0103] Determine the missing identification results based on the manual identification results; when determining the missing identification results in the manual identification results, obtain the historical material identification results of the manipulated materials based on big data, compare the historical material identification results with the manual identification results, and determine the missing identification results that do not exist in the manual identification results;

[0104] Inputting the missing identification result into the grinding parameter determination model, obtaining the missing grinding parameter output by the grinding parameter determination model, characterizing the missing identification result, obtaining a second manipulated material identification result feature value set, and associating it with the missing grinding parameter to obtain a second associated item;

[0105] The first associated items and the second associated items of all the manipulated materials are integrated to obtain a grinding strategy library of the sand mill of the sand mill type.

[0106] The working principle and beneficial effects of the above technical solution are:

[0107] The present invention introduces the operation records of operators of sand mills of different types, obtains manual recognition results and grinding parameters in the operation records, performs machine learning on the manual recognition results and grinding parameters of the same controlled material to obtain a grinding parameter determination model; obtains missing recognition results in the manual recognition results, inputs the missing recognition results into the grinding parameter determination model, and obtains the missing grinding parameters output by the grinding parameter determination model; characterizes the manual recognition results to obtain a first controlled material recognition result feature value set and associates them with the grinding parameters to obtain a first associated item; characterizes the missing recognition results to obtain a second controlled material recognition result feature value set and associates them with the missing grinding parameters to obtain a second associated item; integrates the first associated items and the second associated items of all controlled materials to obtain a grinding strategy library of different types of sand mills. In addition to directly obtaining data items of the grinding strategy library from known operation records, machine learning is performed using known data to supplement the missing grinding parameters of the missing recognition results, thereby improving the comprehensiveness and efficiency of the construction of the grinding strategy library.

[0108] The embodiment of the present invention provides a material type identification system applied to the automated operation of a sand mill, further comprising:

[0109] The verification subsystem is used to obtain the sampled material after the identification material is ground, and to verify the grinding accuracy of the sampled material. If the grinding accuracy verification fails, the grinding medium test is performed. The sampled material is the part of the material taken out after the identification material is ground and used for the grinding accuracy test.

[0110] The working principle and beneficial effects of the above technical solution are:

[0111] The present invention samples the identified material after grinding, and when the grinding accuracy check of the sampled material fails, the grinding medium is tested in time, thereby improving the timeliness of the replacement of the grinding medium.

[0112] In one embodiment, when the grinding accuracy verification fails, the verification subsystem performs a grinding medium test and performs the following operations:

[0113] Obtain real-time triboelectric signals between friction pairs; wherein the friction pair is: the parts of the sand mill where two contact surfaces rub against each other during relative motion, such as: the dynamic ring and static ring of a mechanical seal;

[0114] The triboelectric signal is obtained by connecting one end of a wire to any object that constitutes the friction pair, and the other end of the wire is connected to a detector, which is read by a detector (such as an oscilloscope). The real-time triboelectric signal is read by the detector when the grinding accuracy verification fails.

[0115] Acquire an initial triboelectric signal of the friction pair; wherein the initial triboelectric signal is an electrical signal read by the detector just after the grinding medium is replaced;

[0116] Calculate the change in friction coefficient based on the electrical signal difference between the real-time friction electric signal and the initial friction electric signal; wherein the change in friction coefficient can be characterized by the electrical signal difference, and the corresponding relationship between the electrical signal and the friction coefficient is obtained based on historical test data;

[0117] If the change of the friction coefficient reaches a preset change degree, the corresponding grinding medium is used as the determination grinding medium, and the use time of the determination grinding medium is obtained; wherein the preset change degree is manually preset, and is the difference between the friction coefficient when the grinding requirement is not met and the friction coefficient when the grinding requirement is met; the use time is: the time the determination grinding medium is put into the sand mill and used;

[0118] Determine the rationality of the use time of the grinding media. If it is unreasonable, reset the cleaning time setting parameters;

[0119] Among them, the rationality of the use time of the grinding medium is judged. If it is unreasonable, the cleaning time setting parameters are reset, including:

[0120] Obtaining and determining a standard usage time of the grinding medium; wherein the standard usage time is: an expected usage time determined by a person who determines the grinding medium based on experience;

[0121] If the time difference between the standard usage time and the usage time is greater than or equal to the preset time difference threshold, the cleaning time setting parameter is obtained; wherein the time difference is: the result obtained by subtracting the usage time from the standard usage time; the preset time difference threshold is manually preset; the cleaning time setting parameter is: the preset cleaning time of the sand mill after grinding a material each time;

[0122] Obtain a cleaning time setting parameter downward adjustment value comparison table; wherein the cleaning time setting parameter downward adjustment value comparison table includes: a plurality of one-to-one corresponding time difference values ​​to be matched and cleaning time setting parameter downward adjustment values, the longer the time difference value to be matched, the larger the corresponding downward adjustment value, and the cleaning time setting parameter after downward adjustment based on the maximum cleaning time setting parameter downward adjustment value needs to be greater than a preset cleaning time threshold;

[0123] According to the comparison table of the time difference and the cleaning time setting parameter downward adjustment value, determine the target cleaning time setting parameter after downward adjustment;

[0124] After notifying the staff to replace the determined grinding media, the target cleaning time setting parameters should be applied.

[0125] The working principle and beneficial effects of the above technical solution are:

[0126] The present invention obtains the electrical signal difference between the real-time friction electric signal and the initial friction electric signal, and the change of the electrical signal represents the change of the friction coefficient. When the change of the friction coefficient reaches the level representing that the friction coefficient does not meet the preset expectation, the corresponding grinding medium is used as the judgment grinding medium to obtain the usage time of the judgment grinding medium; the standard usage time and the usage time are compared, and the time difference is obtained by difference. When the time difference is greater than the time difference threshold, it indicates that the wear speed of the grinding medium is too fast. In the gap between the sand mill grinding different materials alternately, the equipment needs to be cleaned. From the perspective of force, when the dispersion disk transfers energy to the grinding medium, if the viscosity of the liquid is very low, the medium may contact the dispersion disk and be broken. Therefore, the cleaning time needs to be shortened as much as possible. Get the cleaning time setting parameters, and also get a comparison table of cleaning time setting parameter downward adjustment values; determine the cleaning time setting parameter downward adjustment value corresponding to the cleaning time setting parameter according to the time difference, and reduce the set cleaning time. At the same time, preset a threshold for the shortest cleaning time to ensure the cleaning effect. After notifying the staff to replace the grinding medium, apply the reduced target cleaning time setting parameters. When the cleaning time setting is unreasonable and causes excessive wear of the grinding medium, promptly notify the staff to replace the grinding medium and reasonably reset the cleaning time setting parameters, thereby avoiding frequent replacement of the grinding medium and being more suitable.

[0127] The embodiment of the present invention provides a material type identification method for sand mill automation operation, such as Figure 2 As shown, including:

[0128] Step 1: Obtain the material type identification result of the material to be ground on the sand mill automation production line;

[0129] Step 2: Retrieve the grinding strategy library according to the material type identification result and determine the grinding strategy for the identified material;

[0130] Step 3: Based on the grinding strategy, control the sand mill to automatically grind the identified material.

[0131] In one embodiment, obtaining the material type identification result of the material to be ground on the sand mill automated production line includes:

[0132] Based on machine vision technology, the image of the material to be ground is obtained and the image features are extracted. According to the image features, the material type recognition result is determined;

[0133] and / or,

[0134] Based on spectral analysis technology, the spectral characteristic information of the material to be ground is analyzed, and then the material type identification result is determined according to the spectral characteristic information;

[0135] and / or,

[0136] The material type recognition model is trained based on the machine learning algorithm and combined with historical data, and then the material type recognition result is determined based on the real-time sensor data and the material type recognition model.

[0137] In one embodiment, the grinding strategy library is retrieved according to the material type identification result to determine the grinding strategy of the identified material, including:

[0138] According to the material type identification result, determine the type of sand mill suitable for grinding the identified material;

[0139] Get the grinding strategy library of the sand mill of the sand mill type;

[0140] Characterize the material type recognition result and obtain the recognition result feature value set;

[0141] The grinding strategy is determined according to the feature value set of the recognition result and the grinding strategy library.

[0142] The embodiment of the present invention provides a material type identification method applied to the automatic operation of a sand mill, and further includes:

[0143] Obtain sampled material after grinding the identified material, perform grinding accuracy verification on the sampled material, and perform grinding medium detection when the grinding accuracy verification fails.

[0144] In one embodiment, when the grinding accuracy check fails, a grinding medium test is performed, including:

[0145] Obtain real-time triboelectric signals between friction pairs;

[0146] Obtaining initial triboelectric signals of the friction pair;

[0147] Calculating the change in friction coefficient according to the electric signal difference between the real-time triboelectric signal and the initial triboelectric signal;

[0148] If the change in the friction coefficient reaches a preset change degree, the corresponding grinding medium is used as the determination grinding medium, and the use time of the determination grinding medium is obtained;

[0149] Determine the rationality of the use time of the grinding media. If it is unreasonable, reset the cleaning time setting parameters;

[0150] Among them, the rationality of the use time of the grinding medium is judged. If it is unreasonable, the cleaning time setting parameters are reset, including:

[0151] Obtain the standard usage time for determining the grinding media;

[0152] If the time difference between the standard usage time and the usage time is greater than or equal to the preset time difference threshold, the cleaning time setting parameter is obtained;

[0153] Obtain a comparison table of cleaning time setting parameter downward adjustment values;

[0154] According to the comparison table of the time difference and the cleaning time setting parameter downward adjustment value, determine the target cleaning time setting parameter after downward adjustment;

[0155] After notifying the staff to change the grinding media, the target cleaning time setting parameters should be applied.

[0156] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A material type identification system used in sand mill automation operation, characterized in that: include: The material type identification subsystem is used to obtain the material type identification result of the material to be ground on the sand mill automated production line; The grinding strategy retrieval subsystem is used to retrieve the grinding strategy library according to the material type identification result and determine the grinding strategy of the identified material; The automatic grinding subsystem is used to control the sand mill to automatically grind the identified materials based on the grinding strategy.

2. A material type identification system for sand mill automation operation as claimed in claim 1, characterized in that: The material type identification subsystem obtains the material type identification result of the material to be ground on the sand mill automated production line and performs the following operations: Based on machine vision technology, the image of the material to be ground is obtained and the image features are extracted. According to the image features, the material type recognition result is determined; and / or, Based on spectral analysis technology, the spectral characteristic information of the material to be ground is analyzed, and then the material type identification result is determined according to the spectral characteristic information; and / or, The material type recognition model is trained based on the machine learning algorithm and combined with historical data, and then the material type recognition result is determined based on the real-time sensor data and the material type recognition model.

3. A material type identification system for sand mill automation operation as claimed in claim 1, characterized in that: The grinding strategy retrieval subsystem retrieves the grinding strategy library according to the material type identification result, determines the grinding strategy of the identified material, and performs the following operations: According to the material type identification result, determine the type of sand mill suitable for grinding the identified material; Get the grinding strategy library of the sand mill of the sand mill type; Characterize the material type recognition result and obtain the recognition result feature value set; The grinding strategy is determined according to the feature value set of the recognition result and the grinding strategy library.

4. A material type identification system for sand mill automation operation as claimed in claim 1, characterized in that: Also perform the following operations: The verification subsystem is used to obtain the sampled material after the identified material is ground, and to verify the grinding accuracy of the sampled material. If the grinding accuracy verification fails, the grinding medium is tested.

5. A material type identification system for sand mill automation operation as claimed in claim 4, characterized in that: Verification subsystem When the grinding accuracy verification fails, perform grinding medium detection and perform the following operations: Obtain real-time triboelectric signals between friction pairs; Obtaining initial triboelectric signals of the friction pair; Calculating the change in friction coefficient according to the electric signal difference between the real-time triboelectric signal and the initial triboelectric signal; If the change in the friction coefficient reaches a preset change degree, the corresponding grinding medium is used as the determination grinding medium, and the use time of the determination grinding medium is obtained; Determine the rationality of the use time of the grinding media. If it is unreasonable, reset the cleaning time setting parameters; Among them, the rationality of the use time of the grinding medium is judged. If it is unreasonable, the cleaning time setting parameters are reset, including: Obtain the standard usage time for determining the grinding media; If the time difference between the standard usage time and the usage time is greater than or equal to the preset time difference threshold, the cleaning time setting parameter is obtained; Obtain a comparison table of cleaning time setting parameter downward adjustment values; According to the comparison table of the time difference and the cleaning time setting parameter downward adjustment value, determine the target cleaning time setting parameter after downward adjustment; After notifying the staff to replace the determined grinding media, the target cleaning time setting parameters should be applied.

6. A material type identification method applied to the automated operation of a sand mill, characterized in that: include: Obtain the material type identification result of the material to be ground on the sand mill automated production line; According to the material type identification result, the grinding strategy library is retrieved to determine the grinding strategy of the identified material; Based on the grinding strategy, the sand mill is controlled to automatically grind the identified materials.

7. A material type identification method for sand mill automation operation as claimed in claim 6, characterized in that: Obtain the material type identification results of the material to be ground on the sand mill automated production line, including: Based on machine vision technology, the image of the material to be ground is obtained and the image features are extracted. According to the image features, the material type recognition result is determined; and / or, Based on spectral analysis technology, the spectral characteristic information of the material to be ground is analyzed, and then the material type identification result is determined according to the spectral characteristic information; and / or, The material type recognition model is trained based on the machine learning algorithm and combined with historical data, and then the material type recognition result is determined based on the real-time sensor data and the material type recognition model.

8. A material type identification method for sand mill automation operation as claimed in claim 6, characterized in that: According to the material type identification result, the grinding strategy library is retrieved to determine the grinding strategy of the identified material, including: According to the material type identification result, determine the type of sand mill suitable for grinding the identified material; Get the grinding strategy library of the sand mill of the sand mill type; Characterize the material type recognition result and obtain the recognition result feature value set; The grinding strategy is determined according to the feature value set of the recognition result and the grinding strategy library.

9. A material type identification method for sand mill automation operation as claimed in claim 6, characterized in that: Also includes: Obtain sampled material after grinding the identified material, perform grinding accuracy verification on the sampled material, and perform grinding medium detection when the grinding accuracy verification fails.

10. A material type identification method for sand mill automation operation as claimed in claim 9, characterized in that: When the grinding accuracy check fails, the grinding media test is performed, including: Obtain real-time triboelectric signals between friction pairs; Obtaining initial triboelectric signals of the friction pair; Calculating the change in friction coefficient according to the electric signal difference between the real-time triboelectric signal and the initial triboelectric signal; If the change in the friction coefficient reaches a preset change degree, the corresponding grinding medium is used as the determination grinding medium, and the use time of the determination grinding medium is obtained; Determine the rationality of the use time of the grinding media. If it is unreasonable, reset the cleaning time setting parameters; Among them, the rationality of the use time of the grinding medium is judged. If it is unreasonable, the cleaning time setting parameters are reset, including: Obtain the standard usage time for determining the grinding media; If the time difference between the standard usage time and the usage time is greater than or equal to the preset time difference threshold, the cleaning time setting parameter is obtained; Obtain a comparison table of cleaning time setting parameter downward adjustment values; According to the comparison table of the time difference and the cleaning time setting parameter downward adjustment value, determine the target cleaning time setting parameter after downward adjustment; After notifying the staff to change the grinding media, the target cleaning time setting parameters should be applied.

Images