Risk prediction system for ore sample preparation process based on artificial intelligence

Abstract

The present invention discloses a risk prediction system for a mineral sample preparation process based on artificial intelligence, and relates to the technical field of mineral sample risk prediction. The present invention collects parameter information composed of process parameter information and key parameter information in the mineral sample preparation process, as well as state information composed of intermediate product state data and equipment state data, and starts from multi-dimensional qualitative and quantitative data of important preparation process parameters, key parameters, intermediate product states, and equipment states, and performs classified parameter clearance analysis and state clearance analysis to output different types of parameter signals and state signals, so as to facilitate subsequent identification and prediction of different types of risks and output analysis strategies for qualified processing conditions, thereby improving the preparation stability and repeatability of ore samples at different collection locations, and improving the accuracy of morphological and component analysis results under qualified processing conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of mineral sample risk prediction, and in particular relates to a risk prediction system for a mineral sample preparation process based on artificial intelligence. Background Art

[0002] At present, in order to ensure the accuracy and repeatability of ore samples, there are many potential risk factors in the ore sample preparation process, such as insufficient technical capabilities of operators, equipment failures, changes in material quality, etc. However, the traditional ore sample preparation process lacks real-time monitoring and risk control measures, resulting in ore sample preparation errors caused by risk factors that are difficult to identify and correct in a timely manner, thereby affecting the accuracy of ore analysis results. There are many types of solid minerals in rocks, such as rock-forming minerals, metallic minerals, non-metallic minerals, etc. Analyzing the types and spatial distribution relationships of various minerals in rocks helps to determine the formation mechanism of these minerals, which plays an important role in mineral prediction.

[0003] There are few reports on risk analysis and prediction systems in the ore sample preparation process in the prior art. Most of them can only realize real-time monitoring and alarm of several key data, and do not collect, analyze and predict risks for classified parameters and states to improve the preparation stability and repeatability of ore samples at different collection locations, thereby further improving the accuracy of the ore sample morphology and component analysis results. Summary of the invention

[0004] The purpose of the present invention is to provide a risk prediction system for the ore sample preparation process based on artificial intelligence, which is used to solve the technical problem that the prior art does not collect, analyze and predict risks based on classified parameters and states, so as to improve the preparation stability and repeatability of ore samples at different collection positions, thereby further improving the accuracy of the ore sample morphology and component analysis results.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] The risk prediction system of the ore sample preparation process based on artificial intelligence includes a parameter acquisition module, a state acquisition module, a risk analysis module, a risk identification and prediction module, an ore sample analysis strategy module and a visual display module; the ore sample preparation process is to collect crude ore from a mine or a mining site, crush the crude ore by a crushing device to obtain ore crushed material, grind the ore crushed material by a grinding device to obtain ore fine powder, sieve the ore fine powder by a sieve to obtain ore sieved powder within a required particle size range, weigh the ore sieved powder by a weighing device to obtain ore weighing powder, dry the ore weighing powder by a drying device to obtain an ore sample, and place the ore sample in a storage container for subsequent analysis;

[0007] The parameter acquisition module is used to collect process parameter information and key parameter information in the ore sample preparation process in real time and send it to the risk analysis module;

[0008] The status acquisition module is used to collect the intermediate product status data and equipment status data in the ore sample preparation process in real time and send it to the risk analysis module;

[0009] The risk analysis module is used to perform parameter clearance analysis on the process parameter information and key parameter information in the ore sample preparation process, output different types of parameter signals and send them to the risk identification and prediction module; it is also used to perform status clearance analysis on the intermediate status data and equipment status data in the ore sample preparation process, output different types of status signals and send them to the risk identification and prediction module.

[0010] Further preferably, the process parameter information includes collection process parameters and processing process parameters, and the key parameter information includes grinding key parameters and screening key parameters; the collection process parameters include the collection location and collection time of the ore sample, and the processing process parameters include the crushing time, grinding time and drying time of the ore sample; the grinding key parameters include the grinding disc speed and the grinding ball medium diameter of the grinding equipment; the screening key parameters include the mesh size of the screen;

[0011] The intermediate product status data includes the ore screening powder status data and the ore sample status data, and the equipment status data includes the grinding equipment data and the weighing equipment data; the ore screening powder status data includes the particle size range of the ore screening powder, and the ore sample status data includes the water content and observed morphology of the ore samples; the grinding equipment data includes the degree of tool wear of the grinding equipment, and the weighing equipment data includes the stability and accuracy of the weighing equipment.

[0012] Further preferably, the specific process of the parameter clearance analysis is as follows:

[0013] S1, sorting out the crushing time, grinding time, drying time of ore samples at different collection locations and different collection times, as well as the grinding disc speed, grinding ball medium diameter, and mesh size of the grinding equipment, calculating the geometric mean of the crushing time, grinding time, and drying time of ore samples at different collection locations and different collection times, naming the geometric mean as the same-collection average value, comparing the same-collection average value with the same-collection average value preset range value, and when the same-collection average value is within the same-collection average value preset range value, entering step S2; when the same-collection average value is not within the same-collection average value preset range value, generating a processing parameter abnormality signal and sending it to the risk identification and prediction module;

[0014] S2. The grinding disc speed, grinding ball medium diameter, and mesh size of the ore samples at different collection locations and collection times are marked as KRct, KJct, and KSct, respectively. c corresponds to different collection locations, and t corresponds to different collection times. According to the formula The comprehensive parameter factor KGct of ore samples collected at different collection locations and times is calculated;

[0015] S3. Compare the passing comprehensive parameter factor with the preset range value of the passing comprehensive parameter factor. When the passing comprehensive parameter factor is within the preset range value of the passing comprehensive parameter factor, generate a normal comprehensive parameter signal and send it to the risk identification and prediction module; when the passing comprehensive parameter factor is not within the preset range value of the passing comprehensive parameter factor, generate a comprehensive parameter risk signal and send it to the risk identification and prediction module.

[0016] Further preferably, the specific process of the status clearance analysis is as follows:

[0017] S11, sorting out the particle size range of the ore screened powder at different collection locations and different collection times, the water content and observed morphology of the ore samples, the tool wear degree of the grinding equipment, and the stability and accuracy of the weighing equipment. When the particle size range of the ore screened powder does not match the mesh size of the screen, generating a screening state abnormality signal and sending it to the risk identification and prediction module; when the particle size range of the ore screened powder matches the mesh size of the screen, entering step S12;

[0018] S12, when the ore sample is observed to be in the form of sticky particles or / and the water content is greater than the water content threshold, a dry state abnormality signal is generated and sent to the risk identification and prediction module; when the ore sample is observed to be in the form of dispersed particles and the water content is less than the water content threshold, step S13 is entered;

[0019] S13, the tool wear area ratio, standard deviation, and absolute error corresponding to the tool wear degree of the grinding equipment, the stability and accuracy of the weighing equipment are marked as SMct, SWct, and SZct, respectively, c corresponds to different collection position areas, t corresponds to different collection times, according to the formula The clearance state factor SGct of ore samples in different collection locations and at different collection times is calculated;

[0020] S14. Compare the pass state factor with the pass state factor threshold. When the pass state factor is less than or equal to the pass state factor threshold, generate a comprehensive normal signal and send it to the risk identification and prediction module. When the pass state factor is greater than the pass state factor threshold, generate a comprehensive risk signal and send it to the risk identification and prediction module.

[0021] Further preferably, the risk identification and prediction module is used to identify and predict as a high-risk condition in the processing process after receiving an abnormal processing parameter signal; identify and predict as a risk condition in the processing process after receiving a comprehensive parameter risk signal; identify and predict as a high-risk condition in the screening process after receiving an abnormal screening state signal; identify and predict as a high-risk condition in the drying process after receiving an abnormal drying state signal; identify and predict as a high-risk condition in the grinding and weighing equipment after receiving a comprehensive state risk signal; and is also used to identify and predict as a qualified processing condition after receiving a comprehensive parameter normal signal and a comprehensive state normal signal.

[0022] Further preferably, the ore sample analysis strategy module is used to provide a matching ore sample analysis strategy for the ore sample identified and predicted to be in a qualified processing condition, and send the matching ore sample analysis strategy to the visualization display module.

[0023] Further preferably, the specific process of providing a matching ore sample analysis strategy for ore samples that are identified and prejudged as being in a qualified processing condition is as follows:

[0024] S21. When the analysis requirement is "macroscopic feature observation, crystal morphology and particle size", the strategy of "ore samples are observed and analyzed through an optical microscope or a magnifying glass" is output;

[0025] S22. When the analysis requirement is "rock composition and microstructure", the strategy of "polishing and polishing, and analysis through optical microscope" is output;

[0026] S23. When the analysis requirement is "microstructure and composition analysis", the strategy of "fixing on a probe sheet and observing and analyzing through a scanning electron microscope or a transmission electron microscope" is output.

[0027] Further preferably, the visualization display module is used to display in real time the collection location area and collection time corresponding to high-risk conditions in the processing process, risk conditions in the processing process, high-risk conditions in the screening process, high-risk conditions in the drying process, high-risk conditions in the grinding and weighing equipment, and qualified processing conditions; and is also used to display in real time subtitles corresponding to the analysis requirements and strategies of ore samples in qualified processing conditions.

[0028] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0029] 1. The present invention collects parameter information consisting of process parameter information and key parameter information in the ore sample preparation process, as well as status information consisting of intermediate product status data and equipment status data, and starts from multi-dimensional qualitative and quantitative data of important preparation process parameters, key parameters, intermediate product status, and equipment status, and performs classified parameter clearance analysis and status clearance analysis to output different types of parameter signals and status signals, which is convenient for subsequent identification and prediction of different types of risks and output analysis strategies for qualified processing conditions, thereby improving the preparation stability and repeatability of ore samples at different collection locations, and improving the accuracy of morphological and component analysis results under qualified processing conditions.

[0030] 2. The risk analysis method of the present invention includes parameter clearance analysis and state clearance analysis. Some important parameters are preferentially selected to reflect special abnormal situations, and then factors reflecting the risk of parameter or state abnormality are obtained by combining marking and weight calculation. The normal or risk signal is obtained by comparing the threshold or the preset range value, which is convenient for subsequent pre-judgment and feedback of risk signals, reduces the risk incidence rate in the ore sample preparation process, and improves the repeatability of ore sample preparation.

[0031] 3. The present invention automatically identifies and predicts specific risk conditions based on the different parameter signals or status signals received, which is convenient for locking the risk occurrence process or equipment, facilitating the location of the fault point and the timeliness of maintenance and inspection, and at the same time facilitating the subsequent analysis and model establishment of ore samples in qualified processing conditions; the analysis strategy is automatically output according to the analysis requirements, which is convenient for taking corresponding analysis measures according to the purpose of ore sample analysis, and improving the efficiency and accuracy of morphology and component analysis of qualified ore samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 The module diagram of the risk prediction system of the ore sample preparation process based on artificial intelligence of the present invention is shown;

[0034] Figure 2 The flowchart of the risk prediction method of the ore sample preparation process based on artificial intelligence of the present invention is shown;

[0035] Figure 3 A subtitle diagram showing the analysis requirements and output strategies of the ore sample analysis strategy module of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] The ore sample preparation process of the present invention is based on the common preparation method in the prior art and specifically includes the following steps:

[0038] Collecting ore samples: Collecting ore samples from mines or mining sites, etc., and recording the location and time of collection;

[0039] Sample crushing: crushing large ore samples into appropriate particle sizes for analytical processing in subsequent steps;

[0040] Sample segmentation: segment the crushed sample into appropriate sizes for subsequent sample preparation;

[0041] Sample grinding: Grind the sample to make it finely powdered and evenly dispersed;

[0042] Sample sieving: Sieve the ground sample to obtain a sample within the required particle size range;

[0043] Sample weighing: Weigh the obtained samples according to the experimental requirements to ensure that the prepared samples meet the experimental requirements;

[0044] Sample mixing: Mix different ore samples in a certain proportion to obtain a more representative composite sample;

[0045] Sample drying: Place the sample in a desiccator to dry it to remove moisture and humidity from the sample;

[0046] Sample storage: Store the prepared samples in appropriate containers and label them for subsequent chemical analysis or other experiments.

[0047] The equipment used in the above preparation process include crusher, grinder, sieve, weigher, dryer and storage container. The crusher is used to crush large pieces of ore samples into appropriate particle size; the grinder is used to grind the sample to make it fine powder and evenly dispersed; the sieve is used to screen the ground sample to obtain the sample within the required particle size range; the weigher is used to weigh the sample accurately; the dryer is used to dry the sample to remove water and moisture from the sample; the storage container is used to store the prepared samples and needs to be airtight and have appropriate capacity.

[0048] Example 1

[0049] See also Figure 1 As shown, a risk prediction system for a mineral sample preparation process based on artificial intelligence in this embodiment includes a parameter acquisition module, a state acquisition module, a risk analysis module, a risk identification and prediction module, a mineral sample analysis strategy module and a visualization display module.

[0050] The preparation process of the ore sample of the present invention is to collect crude ore from a mine or a mining site, crush the crude ore by a crushing device to obtain ore crushed material, grind the ore crushed material by a grinding device to obtain ore fine powder, sieve the ore fine powder by a sieve to obtain ore screened powder within a required particle size range, weigh the ore screened powder by a weighing device to obtain ore weighing powder, dry the ore weighing powder by a drying device to obtain an ore sample, and place the ore sample in a storage container for subsequent analysis.

[0051] The parameter acquisition module is used to collect process parameter information and key parameter information in the ore sample preparation process in real time and send it to the risk analysis module; the process parameter information includes acquisition process parameters and processing process parameters, and the key parameter information includes grinding key parameters and screening key parameters.

[0052] Specifically, the collection process parameters include the collection location and time of the ore samples; the processing process parameters include the crushing time, grinding time and drying time of the ore samples; the key parameters of grinding include the grinding disc speed and the grinding ball medium diameter of the grinding equipment; the key parameters of screening include the mesh size of the screen.

[0053] The collection location is marked by traditional three-dimensional coordinates. After the custom coordinate origin is defined, the collection location of the ore sample is represented by the coordinate system (Xi, Yi, Zi), i=1,2,...n, n corresponds to each batch of collection. The area division of the collection location is centered on the custom coordinate origin, and the area within the range of (Xi±5m, Yi±5m, Zi±5m) is defined as the same collection location area.

[0054] The grinding disc speed directly affects the grinding efficiency and grinding time of the grinding equipment. Too high or too low speed will lead to uneven grinding or over-grinding. The hardness and size of the ball mill medium will affect the grinding effect. Proper selection of grinding balls of suitable material and size can improve the grinding efficiency and grinding quality. The mesh size of the screen directly affects the particle size distribution of the sample.

[0055] The status acquisition module is used to collect the intermediate product status data and equipment status data in the ore sample preparation process in real time and send it to the risk analysis module; the intermediate product status data includes ore screening powder status data and ore sample status data, and the equipment status data includes grinding equipment data and weighing equipment data.

[0056] Specifically, the ore screened powder state data includes the particle size range of the ore screened powder, the ore sample state data includes the water content and observed morphology of the ore sample; the grinding equipment data includes the degree of tool wear of the grinding equipment, and the weighing equipment data includes the stability and accuracy of the weighing equipment.

[0057] The moisture content of ore samples can directly reflect the water and moisture conditions of ore samples, and indirectly reflect the drying effect of drying equipment; the observed forms of ore samples include dispersed particles and sticky particles.

[0058] The degree of tool wear of the above-mentioned grinding equipment refers to the proportion of tool wear area, which is obtained by dividing the wear area of ​​the tool by the total area of ​​the tool. The wear area is calculated through pixel points on the image with the help of image processing and computer vision technology.

[0059] The stability of a weighing device refers to the degree of fluctuation in the readings when objects of the same mass are weighed continuously, expressed as standard deviation. A reading fluctuation with a standard deviation of less than 0.1% is judged to have good stability. The accuracy of a weighing device refers to the difference between the reading of the weighing device and the actual mass, determined by calibration testing and measured by absolute error. An absolute error of less than 0.1% is judged to have good accuracy.

[0060] The risk analysis module is used to perform parameter clearance analysis on the process parameter information and key parameter information in the ore sample preparation process, output different types of parameter signals and send them to the risk identification and prediction module; it is also used to perform status clearance analysis on the intermediate status data and equipment status data in the ore sample preparation process, output different types of status signals and send them to the risk identification and prediction module.

[0061] This embodiment collects parameter information consisting of process parameter information and key parameter information in the ore sample preparation process, as well as status information consisting of intermediate product status data and equipment status data, and starts from multi-dimensional qualitative and quantitative data of important preparation process parameters, key parameters, intermediate product status, and equipment status, and performs classified parameter clearance analysis and status clearance analysis to output different types of parameter signals and status signals, which is convenient for subsequent identification and prediction of different types of risks and output analysis strategies for processing qualified conditions, thereby improving the preparation stability and repeatability of ore samples at different collection locations, and improving the accuracy of morphological and component analysis results under processing qualified conditions.

[0062] Example 2

[0063] See also Figure 1-Figure 2As shown, the risk prediction system for the ore sample preparation process based on artificial intelligence in this embodiment, based on Example 1, further discloses the specific process of parameter clearance analysis and state clearance analysis, and discloses the functions of the risk identification and prediction module and the automatic feedback adjustment module.

[0064] Specifically, the specific process of parameter clearance analysis is as follows:

[0065] S1, sorting out the crushing time, grinding time, drying time of ore samples at different collection locations and different collection times, as well as the grinding disc speed, grinding ball medium diameter, and mesh size of the grinding equipment, calculating the geometric mean of the crushing time, grinding time, and drying time of ore samples at different collection locations and different collection times, naming the geometric mean as the same-collection average value, comparing the same-collection average value with the same-collection average value preset range value, and when the same-collection average value is within the same-collection average value preset range value, entering step S2; when the same-collection average value is not within the same-collection average value preset range value, generating a processing parameter abnormality signal and sending it to the risk identification and prediction module;

[0066] S2. The grinding disc speed, grinding ball medium diameter, and mesh size of the ore samples at different collection locations and collection times are marked as KRct, KJct, and KSct, respectively. c corresponds to different collection locations, and t corresponds to different collection times. According to the formula The comprehensive parameter factor KGct of ore samples collected at different collection locations and times is calculated; wherein w1, w2, and w3 are all preset weight coefficients and 2.37>w1>w2>w3>1.06;

[0067] S3. Compare the passing comprehensive parameter factor with the preset range value of the passing comprehensive parameter factor. When the passing comprehensive parameter factor is within the preset range value of the passing comprehensive parameter factor, generate a normal comprehensive parameter signal and send it to the risk identification and prediction module; when the passing comprehensive parameter factor is not within the preset range value of the passing comprehensive parameter factor, generate a comprehensive parameter risk signal and send it to the risk identification and prediction module.

[0068] When the apparent value of the passing comprehensive parameter factor is not within the preset range of the passing comprehensive parameter factor, it means that the probability of abnormal risk in the comprehensive situation of the process and key parameters corresponding to the collection time in the collection location area is greater, and the negative impact on the final morphology and analysis results of the ore sample is greater.

[0069] The parameter passing analysis process first calculates the same-mining average value of the three processing stages for ore samples in different collection locations and at different collection times. The same-mining average value reflects the comprehensive abnormal conditions of the three processing stages, and the comparison with the preset range value can prioritize the abnormal conditions of the processing process; then, other parameters are marked and weighted to obtain the passing comprehensive parameter factor that reflects the comprehensive abnormal risk of the parameters, and the preset range value comparison method is combined to generate a signal of whether the comprehensive parameter is normal or risky, which is convenient for the subsequent prediction and feedback of the comprehensive parameter risk.

[0070] The specific process of state clearance analysis is as follows:

[0071] S11, sorting out the particle size range of the ore screened powder at different collection locations and different collection times, the water content and observed morphology of the ore samples, the tool wear degree of the grinding equipment, and the stability and accuracy of the weighing equipment. When the particle size range of the ore screened powder does not match the mesh size of the screen, generating a screening state abnormality signal and sending it to the risk identification and prediction module; when the particle size range of the ore screened powder matches the mesh size of the screen, entering step S12;

[0072] S12, when the ore sample is observed to be in the form of sticky particles or / and the water content is greater than the water content threshold, a dry state abnormality signal is generated and sent to the risk identification and prediction module; when the ore sample is observed to be in the form of dispersed particles and the water content is less than the water content threshold, step S13 is entered;

[0073] S13, the tool wear area ratio, standard deviation, and absolute error corresponding to the tool wear degree of the grinding equipment, the stability and accuracy of the weighing equipment are marked as SMct, SWct, and SZct, respectively, c corresponds to different collection position areas, t corresponds to different collection times, according to the formula The clearance status factor SGct of ore samples at different collection locations and times is calculated; wherein s1, s2, and s3 are all preset weight coefficients and 2.16>s3>s2>s1>1.13;

[0074] S14. Compare the pass state factor with the pass state factor threshold. When the pass state factor is less than or equal to the pass state factor threshold, generate a comprehensive normal signal and send it to the risk identification and prediction module. When the pass state factor is greater than the pass state factor threshold, generate a comprehensive risk signal and send it to the risk identification and prediction module.

[0075] The larger the apparent value of the pass status factor, the greater the probability that the equipment status at the corresponding collection time in the collection location area will have an abnormal risk, and the greater the negative impact on the final form and analysis results of the ore sample.

[0076] The status pass analysis first matches the particle size range of the ore screened powder with the mesh size of the screen, giving priority to reflecting whether the screening status is abnormal; then the observed morphology and moisture content of the intermediate product status are combined to reflect whether the drying state of the ore sample is abnormal; then the equipment status data is marked and weighted to obtain a pass status factor that reflects the comprehensive abnormal risk of the status, and combined with the threshold comparison method to generate a signal of overall normal or risky status, which is convenient for subsequent prediction and feedback of the overall risk.

[0077] The risk identification and prediction module is used to identify and predict that the processing process is in a high-risk condition after receiving an abnormal processing parameter signal; identify and predict that it is a risk condition in the processing process after receiving a comprehensive parameter risk signal; identify and predict that it is a high-risk condition in the screening process after receiving an abnormal screening state signal; identify and predict that it is a high-risk condition in the drying process after receiving an abnormal drying state signal; identify and predict that it is a high-risk condition in the grinding and weighing equipment after receiving a comprehensive state risk signal; and is also used to identify and predict that it is a qualified processing condition after receiving a comprehensive parameter normal signal and a comprehensive state normal signal.

[0078] The risk identification and prediction module automatically identifies and predicts specific risk conditions based on the different parameter signals or status signals received, which facilitates the locking of risk-occurring processes or equipment, facilitates the location of fault points and the timeliness of maintenance and inspection, and facilitates subsequent analysis and model building of ore samples in qualified processing conditions.

[0079] The ore sample analysis strategy module is used to provide a matching ore sample analysis strategy for ore samples that are identified and predicted to be qualified for processing, and send the matching ore sample analysis strategy to the visualization display module.

[0080] See also Figure 3 As shown in the figure, the specific process of providing a matching ore sample analysis strategy for identifying ore samples that are predicted to be qualified for processing is as follows:

[0081] S21. When the analysis requirement is "macroscopic feature observation, crystal morphology and particle size", the strategy of "ore samples are observed and analyzed through an optical microscope or a magnifying glass" is output;

[0082] S22. When the analysis requirement is "rock composition and microstructure", the strategy of "polishing and polishing, and analysis through optical microscope" is output;

[0083] S23. When the analysis requirement is "microstructure and composition analysis", the strategy of "fixing on a probe sheet and observing and analyzing through a scanning electron microscope or a transmission electron microscope" is output.

[0084] The ore sample analysis strategy can automatically output the analysis strategy according to the analysis requirements, making it easy to take corresponding analysis measures according to the purpose of ore sample analysis, and improving the efficiency and accuracy of processing qualified ore sample morphology and component analysis.

[0085] The visualization display module is used to display in real time the collection location area and collection time corresponding to high-risk conditions in the processing process, risk conditions in the processing process, high-risk conditions in the screening process, high-risk conditions in the drying process, high-risk conditions in the grinding and weighing equipment, and qualified processing conditions; it is also used to display in real time the subtitles corresponding to the analysis requirements and strategies of ore samples in qualified processing conditions.

[0086] The preset weight coefficients mentioned in this application are used to balance the weights of each data in the formula calculation, so as to promote the accuracy of the calculation results; the size of the coefficient is to quantify each parameter to obtain a specific value, which is convenient for subsequent comparison. The size of the coefficient depends on the amount of sample data and the preliminary setting of the corresponding weight factor coefficient for each group of sample data by technical personnel in this field; as long as it does not affect the proportional relationship between the parameter and the quantized value. The formulas involved are all obtained by collecting a large amount of data, removing dimensions, and performing software simulation, and a formula close to the true value is selected. The coefficients in the formula are set by technical personnel in this field according to actual conditions. The thresholds and preset range values ​​mentioned in this application are obtained and selected by software communication technicians after collecting a large amount of data for software simulation.

[0087] Example 3

[0088] See also Figure 2 As shown, this embodiment provides a risk prediction method for a mineral sample preparation process based on artificial intelligence, comprising the following steps:

[0089] S100, parameter and status collection: real-time collection of process parameter information and key parameter information in the ore sample preparation process, real-time collection of intermediate product status data and equipment status data in the ore sample preparation process;

[0090] S200, risk analysis: perform parameter clearance analysis on process parameter information and key parameter information in the ore sample preparation process, and output different types of parameter signals; perform status clearance analysis on intermediate product status data and equipment status data in the ore sample preparation process, and output different types of status signals; the above different types of parameter signals include processing parameter abnormality signals, comprehensive parameter risk signals and comprehensive parameter normal signals, and the above different types of status signals include screening status abnormality signals, drying status abnormality signals, comprehensive state normal signals and comprehensive state risk signals;

[0091] S300, risk identification and prejudgment: after receiving a processing parameter abnormal signal, identify and prejudge it as a high-risk condition in the processing process; after receiving a comprehensive parameter risk signal, identify and prejudge it as a risk condition in the processing process; after receiving a screening state abnormal signal, identify and prejudge it as a high-risk condition in the screening process; after receiving a drying state abnormal signal, identify and prejudge it as a high-risk condition in the drying process; after receiving a comprehensive state risk signal, identify and prejudge it as a high-risk condition in the grinding and weighing equipment; it is also used for identifying and prejudging it as a qualified processing condition after receiving a comprehensive parameter normal signal and a comprehensive state normal signal;

[0092] S400, ore sample analysis strategy selection: providing a matching ore sample analysis strategy for the ore sample identified and predicted to be in a qualified processing condition, and sending the matching ore sample analysis strategy to the visualization display module;

[0093] S500, Visual display: Real-time display of the collection location area and collection time corresponding to high-risk conditions in the processing process, risk conditions in the processing process, high-risk conditions in the screening process, high-risk conditions in the drying process, high-risk conditions in the grinding and weighing equipment, and qualified processing conditions; it is also used to display subtitles corresponding to the analysis requirements and strategies of ore samples in qualified processing conditions in real time.

[0094] The risk prediction method for the ore sample preparation process of the present embodiment comprises the steps of parameter and status collection, risk analysis, risk identification and prediction, ore sample analysis strategy selection and visual display. Through classified parameter clearance analysis and status clearance analysis, different types of parameter signals and status signals are output, which is convenient for subsequent identification and prediction of different types of risks and output of analysis strategies for qualified processing conditions, thereby improving the preparation stability and repeatability of ore samples at different collection positions, and improving the accuracy of morphological and component analysis results under qualified processing conditions; automatically outputting analysis strategies according to user analysis requirements is convenient for taking corresponding analysis measures according to the purpose of ore sample analysis, and improving the efficiency and accuracy of morphological and component analysis of qualified ore samples.

[0095] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0096] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. Risk prediction system for ore sample preparation process based on artificial intelligence, It is characterized in that It includes a parameter collection module, a state collection module, a risk analysis module, a risk identification and prediction module, a mineral sample analysis strategy module and a visual display module; the mineral sample preparation process is to collect crude ore from a mine or a mining site, crush the crude ore by a crushing device to obtain ore crushed material, grind the ore crushed material by a grinding device to obtain ore fine powder, sieve the ore fine powder by a sieve to obtain ore sieved powder within the required particle size range, weigh the ore sieved powder by a weighing device to obtain ore weighing powder, dry the ore weighing powder by a drying device to obtain an ore sample, and place the ore sample in a storage container for subsequent analysis; The parameter acquisition module is used to collect process parameter information and key parameter information in the ore sample preparation process in real time and send it to the risk analysis module; The status acquisition module is used to collect the intermediate product status data and equipment status data in the ore sample preparation process in real time and send it to the risk analysis module; The risk analysis module is used to perform parameter clearance analysis on the process parameter information and key parameter information in the ore sample preparation process, output different types of parameter signals and send them to the risk identification and prediction module; It is also used to perform status clearance analysis on the intermediate product status data and equipment status data in the ore sample preparation process, output different types of status signals and send them to the risk identification and prediction module; The process parameter information includes collection process parameters and processing process parameters, and the key parameter information includes grinding key parameters and screening key parameters; the collection process parameters include the collection location and collection time of the ore sample, and the processing process parameters include the crushing time, grinding time and drying time of the ore sample; the grinding key parameters include the grinding disc speed and the grinding ball medium diameter of the grinding equipment; the screening key parameters include the mesh size of the screen; The intermediate product status data includes the ore screening powder status data and ore sample status data, and the equipment status data includes the grinding equipment data and weighing equipment data; the ore screening powder status data includes the particle size range of the ore screening powder, and the ore sample status data includes the water content and observed morphology of the ore sample; the grinding equipment data includes the tool wear degree of the grinding equipment, and the weighing equipment data includes the stability and accuracy of the weighing equipment; The specific process of the parameter clearance analysis is as follows: S1, sorting out the crushing time, grinding time, drying time of ore samples at different collection locations and different collection times, as well as the grinding disc speed, grinding ball medium diameter, and mesh size of the grinding equipment, calculating the geometric mean of the crushing time, grinding time, and drying time of ore samples at different collection locations and different collection times, naming the geometric mean as the same-collection average value, comparing the same-collection average value with the same-collection average value preset range value, and when the same-collection average value is within the same-collection average value preset range value, entering step S2; when the same-collection average value is not within the same-collection average value preset range value, generating a processing parameter abnormality signal and sending it to the risk identification and prediction module; S2. The grinding disc speed, grinding ball medium diameter, and mesh size of the ore samples at different collection locations and collection times are marked as KRct, KJct, and KSct, respectively. c corresponds to different collection locations, and t corresponds to different collection times. According to the formula The comprehensive parameter factor KGct of ore samples collected at different collection locations and times is calculated; wherein w1, w2, and w3 are all preset weight coefficients and 2.37>w1>w2>w3>1.06; S3. Compare the comprehensive parameter factor of the clearance with the preset range value of the comprehensive parameter factor of the clearance. When the comprehensive parameter factor of the clearance is within the preset range value of the comprehensive parameter factor of the clearance, generate a normal comprehensive parameter signal and send it to the risk identification and prediction module; when the comprehensive parameter factor of the clearance is not within the preset range value of the comprehensive parameter factor of the clearance, generate a comprehensive parameter risk signal and send it to the risk identification and prediction module; The specific process of the state clearance analysis is as follows: S11, sorting out the particle size range of the ore screened powder at different collection locations and different collection times, the water content and observed morphology of the ore samples, the tool wear degree of the grinding equipment, and the stability and accuracy of the weighing equipment. When the particle size range of the ore screened powder does not match the mesh size of the screen, generating a screening state abnormality signal and sending it to the risk identification and prediction module; when the particle size range of the ore screened powder matches the mesh size of the screen, entering step S12; S12, when the ore sample is observed to be in the form of sticky particles or / and the water content is greater than the water content threshold, a dry state abnormality signal is generated and sent to the risk identification and prediction module; when the ore sample is observed to be in the form of dispersed particles and the water content is less than the water content threshold, step S13 is entered; S13, the tool wear area ratio, standard deviation, and absolute error corresponding to the tool wear degree of the grinding equipment, the stability and accuracy of the weighing equipment are marked as SMct, SWct, and SZct, respectively, c corresponds to different collection position areas, t corresponds to different collection times, according to the formula The clearance status factor SGct of ore samples at different collection locations and times is calculated; wherein s1, s2, and s3 are all preset weight coefficients and 2.16>s3>s2>s1>1.13; S14. Compare the pass state factor with the pass state factor threshold. When the pass state factor is less than or equal to the pass state factor threshold, generate a comprehensive normal signal and send it to the risk identification and prediction module. When the pass state factor is greater than the pass state factor threshold, generate a comprehensive risk signal and send it to the risk identification and prediction module.

2. The risk prediction system for the ore sample preparation process based on artificial intelligence according to claim 1, It is characterized in that The risk identification and prediction module is used to identify and predict the high-risk condition of the processing process after receiving the abnormal processing parameter signal; identify and predict the risk condition of the processing process after receiving the comprehensive parameter risk signal; identify and predict the high-risk condition of the screening process after receiving the abnormal screening state signal; identify and predict the high-risk condition of the drying process after receiving the abnormal drying state signal; After receiving the comprehensive risk signal, it is identified and predicted as a high-risk condition of the grinding and weighing equipment; It is also used to identify and predict the processing as qualified after receiving the normal comprehensive parameter signal and the normal comprehensive status signal.

3. The risk prediction system for the ore sample preparation process based on artificial intelligence according to claim 1, It is characterized in that The ore sample analysis strategy module is used to provide a matching ore sample analysis strategy for the ore sample that is identified and predicted to be in a qualified processing condition, and send the matching ore sample analysis strategy to the visualization display module.

4. The risk prediction system for the ore sample preparation process based on artificial intelligence according to claim 3, It is characterized in that The specific process of providing a matching ore sample analysis strategy for identifying ore samples that are predicted to be qualified for processing is as follows: S21. When the analysis requirement is "macroscopic feature observation, crystal morphology and particle size", the strategy of "ore samples are observed and analyzed through an optical microscope or a magnifying glass" is output; S22. When the analysis requirement is "rock composition and microstructure", the strategy of "polishing and polishing, and analysis through optical microscope" is output; S23. When the analysis requirement is "microstructure and composition analysis", the strategy of "fixing on a probe sheet and observing and analyzing through a scanning electron microscope or a transmission electron microscope" is output.

5. The risk prediction system for the ore sample preparation process based on artificial intelligence according to claim 1, It is characterized in that The visualization display module is used to display in real time the collection location area and collection time corresponding to the high-risk conditions of the processing process, the risk conditions in the processing process, the high-risk conditions of the screening process, the high-risk conditions of the drying process, the high-risk conditions of the grinding and weighing equipment, and the qualified processing conditions; It is also used to display in real time the subtitles corresponding to the analysis requirements and strategies of ore samples in qualified processing status.

Images