High-carbon ferrochrome tailings multi-stage crushing recovery magnetic separation control method

Abstract

The application belongs to the technical field of magnetic separation control, and specifically discloses a high-carbon ferrochrome tailing multi-stage crushing and recovery magnetic separation control method. The method realizes forward control of nodes by linking multi-stage crushing and multi-stage magnetic separation processes. Specifically, the method comprises the following steps: judging the change requirement of each stage of crushing process based on the real-time index of the magnetic separation process; dynamically setting the discharge particle size reference range for the process with the change requirement; judging whether to trigger the crushing equipment adjustment by comparing the real-time discharge particle size with the reference range; when the tailing grade shows an abnormal upward trend, further determining the abnormal responsibility subject based on the real-time feed grade. Finally, according to the classified disposal of the responsibility subject, the application effectively changes the after-the-fact remedy into the before-the-fact prevention, significantly reduces the metal loss risk, improves the resource recovery efficiency and the stability of the control effect, and improves the overall control efficiency and control effect of the complex working condition.

Description

Technical Field

[0001] This invention belongs to the field of magnetic separation control technology, and specifically relates to a magnetic separation control method for multi-stage crushing and recycling of high-carbon ferrochrome tailings. Background Technology

[0002] The production of high-carbon ferrochrome generates a large amount of tailings. If these tailings are not effectively recycled, they will not only waste resources but also put significant pressure on the environment. As the importance of comprehensive resource utilization and environmental protection continues to increase, the requirements for tailings recycling and control are also becoming more stringent.

[0003] The prior art 1, as disclosed in Chinese invention patent application No. 202311072285.5, is a tailings control method and terminal based on AI technology for washing magnetic separators. This method uses AI vision technology to identify overflow products in real time and determine the tailings grade online, thereby dynamically adjusting the operating parameters of the washing magnetic separator, such as water supply and magnetic field strength, to achieve precise and efficient closed-loop control of tailings grade.

[0004] Prior art 2, such as the Chinese invention patent application with application number 202311406610.7, discloses a magnetic separation optimization control method based on knowledge graph. It constructs a magnetic separation knowledge graph reasoning model and integrates reinforcement learning strategies to explore unknown strategies based on learning from historical control experience, thereby achieving coordinated optimization and closed-loop control of the feed rate and magnetic field strength of multiple magnetic separators.

[0005] Existing technology 1 relies on tailings image-based back-induction control, but it suffers from a lag. When tailings grade is abnormal, the loss has already occurred, making it a reactive measure. Existing technology 2 relies on historical data, which may not be effective in dealing with new operating conditions. Obviously, current magnetic separation control for recovery only starts with data after the magnetic separation process. By the time a problem is detected, the substandard feed has already entered the magnetic separator. The control behavior is essentially remedial rather than preventative, resulting in a lack of significant improvement in control effectiveness. It cannot achieve greater efficiency optimization by changing the fundamental factor of feed properties. At the same time, the control strategies of both existing technologies assume that production is in a relatively stable state. When there are sudden changes in incoming materials, there is a lack of fundamental response strategies, making effective control impossible. Summary of the Invention

[0006] In view of this, in order to solve the above problems, a multi-stage crushing, recycling, and magnetic separation control method for high-carbon ferrochrome tailings is proposed.

[0007] The objective of this invention can be achieved through the following technical solution: This invention provides a multi-stage crushing, recovery, and magnetic separation control method for high-carbon ferrochrome tailings. The method includes: dynamically setting the discharge particle size reference range for each stage of the crushing process based on historical discharge records, real-time feed flow rate of each stage of the magnetic separation process, and real-time total recovery rate.

[0008] Based on the aforementioned discharge particle size reference range, combined with real-time concentrate grade and real-time tailings grade, the adjustment trigger status of each stage of crushing equipment and the upward trend of tailings grade are determined in parallel.

[0009] If the tailings grade shows an abnormal upward trend, the responsible party for the abnormality will be determined based on the real-time feed grade of the magnetic separation process. If it only triggers the adjustment of the crushing equipment, the crushing equipment will be adjusted.

[0010] If the abnormality is caused by fluctuations in ore grade, the crushing equipment will be adjusted; if the abnormality is caused by other factors, a collaborative adjustment strategy will be generated and executed.

[0011] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The present invention realizes online evaluation of the operation status of the crushing process and judgment of change requirements by linking the multi-stage magnetic separation process with the multi-stage crushing process, so that the control node is moved from the magnetic separation stage to the crushing stage. Thus, intervention measures are taken before the problematic material enters the magnetic separator. This fundamentally changes the passive situation of post-event remediation, effectively reduces the risk of metal loss caused by unqualified ore feed, and further improves the magnetic separation control effect.

[0012] (2) By dynamically setting the discharge particle size reference range, the present invention enables the crushing process to automatically adjust the discharge standard according to the real-time working conditions, effectively improving the crushing's adaptability to changes in the properties of the incoming material, ensuring that the crushing process always provides materials with stable particle size characteristics for subsequent magnetic separation operations, improving the stability of the feed quality of the magnetic separation process, and realizing precise control of the particle size characteristics of the crushed product, creating favorable conditions for improving the recovery rate and concentrate grade of subsequent magnetic separation processes.

[0013] (3) By determining whether the crushing equipment adjustment is triggered, and when triggered, by determining whether the tailings grade shows an abnormal upward trend and identifying the responsible party for the abnormality, the present invention achieves precise classification and handling of feed grade fluctuations and other process abnormalities, avoids ineffective equipment parameter adjustments, and effectively improves the overall control efficiency of complex working conditions.

[0014] (4) This invention achieves precise control of particle size parameters in the current process by adjusting the crushing equipment. At the same time, through the coordinated adjustment of crushing and magnetic separation equipment, it ensures that the output quality of each crushing stage meets the magnetic separation requirements, and further optimizes the control effect by adjusting the parameters of multiple equipment in complex working conditions. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the overall implementation process of the present invention.

[0017] Figure 2 This is a schematic diagram illustrating the specific implementation process of the modified procedures in this invention.

[0018] Figure 3 This is a schematic diagram illustrating the process of dynamically setting the granulation reference range for changing the process in this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This embodiment describes a practical application scenario integrating three-stage crushing and three-stage magnetic separation. For details, please refer to [link / reference]. Figure 1 As shown, the present invention provides a multi-stage crushing, recovery, and magnetic separation control method for high-carbon ferrochrome tailings. The method includes: S1, dynamically setting the discharge particle size reference range for each stage of the crushing process based on historical discharge records, real-time feed flow rate of each stage of the magnetic separation process, and real-time total recovery rate.

[0021] In one specific embodiment, the crushing process employs a combination of jaw crushers, cone crushers, and sand-making crushers. The primary crushing stage uses jaw crushers to initially crush large pieces of tailings, laying the foundation for subsequent crushing processes. The secondary crushing stage uses cone crushers to further reduce the particle size of the material. The sand-making crusher, as the tertiary crushing equipment, is used to further control the particle size of the material, facilitating the subsequent magnetic separation process.

[0022] To improve the recovery rate of target elements in tailings, this invention employs multiple magnetic separators of different specifications for combined magnetic separation. In one specific embodiment, the primary magnetic separation process uses a 600x950mm magnetic separator to coarsely separate the initially crushed material, removing most of the non-magnetic impurities. The secondary magnetic separation process uses an 800x950mm magnetic separator to further screen for useful elements, improving recovery accuracy. The tertiary magnetic separation process again uses a 600x950mm magnetic separator for deep purification of the material, ensuring full recovery of useful elements.

[0023] In the treatment of high-carbon ferrochrome tailings, the performance of the magnetic separation process is directly constrained by the particle size of the output from the crushing process. Real-time feed flow rate, total recovery rate, concentrate grade, and tailings grade are key indicators of the magnetic separation process, reflecting the particle size distribution and quality of the crushed material. Abnormal changes in these indicators may indicate that the output particle size from the crushing process does not meet requirements, thus affecting the magnetic separation efficiency.

[0024] The adjustment of the crushing process is not based solely on instantaneous data. The impact of changes in the output particle size of the crushing equipment on magnetic separation has a certain delay and cumulative effect. Therefore, it is necessary to locate the root cause of the upstream crushing process from the final performance of the magnetic separation process, thereby determining the crushing process that needs to change the output particle size reference range and setting a targeted output particle size reference range.

[0025] Based on this, in this embodiment of the invention, the crushing process that needs to be changed is determined by combining historical stable intervals, historical average levels, and time windows, i.e., the change process is determined.

[0026] Please see Figure 2 As shown, in one specific embodiment, determining the specific implementation process of the change procedure includes the following steps: First, continuously collect the real-time feed flow rate from the primary magnetic separator drum and calculate the real-time total recovery rate. Then, using a preset duration, such as 30 minutes, calculate the standard deviation of the feed flow rate and total recovery rate within the current time window.

[0027] The calculated standard deviation is compared with the historical stable range of the parameter obtained under normal production conditions. If the standard deviation of the feed flow rate or total recovery rate exceeds the historical stable range, and at the same time, the standard deviation of the coarse particle size range of the primary jaw crusher also exceeds its own historical stable range, then the instability of the feed is highly correlated with the instability of the particle size range of the primary crusher output. In this case, the primary crushing process is set as a change process. The coarse particle size range refers to the difference in particle size between two preset key particle size points in the output of the primary crushing process.

[0028] It should be noted that the two preset critical particle size points are denoted as the first critical particle size point and the second critical particle size point, respectively. The first critical particle size point is... This refers to the particle size value corresponding to a cumulative particle size distribution percentage of 90%. The second critical particle size point is... This refers to the particle size value corresponding to a cumulative particle size distribution percentage of 10%. The coarse-grained range is... and The difference.

[0029] It should also be noted that the historical stability interval can preferably be obtained based on the statistical data of the past 6 months of production data, taking the 95% confidence interval of the fluctuation range of parameters such as feed flow under normal production conditions as its corresponding historical stability interval.

[0030] Subsequently, for the real-time concentrate grade and real-time tailings grade, which better reflect product quality and metal recovery rate, the real-time concentrate grade and real-time tailings grade are divided into multiple continuous analysis time windows based on a preset fixed time interval, such as one data point every 5 minutes.

[0031] When a continuous decline in concentrate grade is detected, for example, when three consecutive analysis time windows show a sustained decrease and the value is below the historical average, this series of consecutive declining windows is marked as an abnormal concentrate grade time window. When a continuous increase in tailings grade is detected and the value is above the historical average, this series of consecutive increasing windows is marked as an abnormal tailings grade time window.

[0032] To address the continuous decline in concentrate grade, within the abnormal concentrate grade time window, Pearson correlation coefficients were calculated between concentrate grade and the main particle concentration of the intermediate cone crusher output, as well as between concentrate grade and the effective particle size ratio of the final sand making crusher output. The process with the strongest positive correlation between concentrate grade and its particle size parameter was identified as the primary responsible process for the decline in concentrate grade and designated as the change process. For example, if the strongest positive correlation exists between concentrate grade and main particle concentration, it indicates that insufficient qualified product output from the cone crusher is the main cause of the product quality decline, thus the intermediate crushing process was designated as the change process.

[0033] It should be noted that the main particle concentration refers to the percentage by weight of particles whose particle size falls within a preset main particle size distribution range. This preset main particle size distribution range is the optimal particle size range for intermediate crushing output determined by back-calculation based on historical best magnetic separation performance; for example, it can be set to 10mm to 30mm. The effective particle size percentage refers to the percentage by weight of particles whose particle size falls within the particle size range that meets the requirements of the magnetic separation process. This particle size range that meets the requirements of the magnetic separation process is a particle size range with both excellent recovery rate and concentrate grade determined through magnetic separation experiments.

[0034] To address the continuous rise in tailings grade, within the abnormal tailings grade time window, Pearson correlation coefficients were calculated between tailings grade and the main particle concentration of the intermediate cone crusher output, as well as between tailings grade and the effective particle size ratio of the final crushing plant output. The crushing process with the strongest negative correlation to tailings grade was identified as the primary responsible process for metal loss and increased tailings grade, and designated as a change process. For example, if the strongest negative correlation exists between tailings grade and the effective particle size ratio, it indicates insufficient effective particles produced by the crushing plant, which is the main cause of inadequate recovery and high tailings loss; thus, the final crushing process would be designated as a change process.

[0035] It should be noted that the Pearson correlation coefficient uses the standard formula, which will not be elaborated further. When the correlation coefficient is positive, the closer the correlation coefficient is to 1, the stronger the positive correlation. When the correlation coefficient is negative, the closer the correlation coefficient is to -1, the stronger the negative correlation.

[0036] It is also important to note that for non-modified processes, the initial discharge particle size reference range should be used as the final discharge particle size reference range. The initial discharge particle size reference range can be manually imported by production personnel or the default setting range for tailings recovery can be used.

[0037] In another specific embodiment of the present invention, please refer to Figure 3 As shown, the specific implementation process for dynamically setting the discharge particle size reference range is as follows: If the process is changed to primary crushing, the goal is to restore the stability of the feed. This involves filtering historical records within a preset critical time window, for example, all historical production records from the last 30 days, and marking these records as candidate historical records. Then, the standard deviation of the feed flow rate and the standard deviation of the total recovery rate within the current time window are obtained. The coefficient of variation of the feed flow rate and the coefficient of variation of the total recovery rate are calculated respectively. The linear weighted sum of these two coefficients is used as the load fluctuation index. Based on the standard deviation of the feed flow rate and the standard deviation of the total recovery rate calibrated for each candidate historical record, the load fluctuation index corresponding to each candidate historical record is calculated. The historical record with the smallest load fluctuation index is then selected from all candidate records. Finally, the coarse particle size range corresponding to the primary crushing process at that time is read from this historical record and directly used as the new discharge particle size reference range for the current primary crushing process.

[0038] It should be noted that the total recovery rate is the final result of a series of processes including crushing, grinding, and magnetic separation. A larger coefficient of variation indicates greater volatility, and fluctuations in the total recovery rate directly indicate uncontrollable changes in output particle size, magnetic field strength, etc. Therefore, the weight of the coefficient of variation of the total recovery rate is set greater than that of the coefficient of variation of the feed flow rate, and their sum is 1. As a preferred example, the weights of the coefficient of variation of the feed flow rate and the coefficient of variation of the total recovery rate can be set to 0.4 and 0.6, respectively. It should also be noted that the weights can be adjusted within a reasonable range according to the specific process characteristics and optimization objectives of different production lines.

[0039] If the process is changed to an intermediate crushing process, the goal is to improve the concentrate grade. That is, based on a large amount of historical data, a system is established with concentrate grade as the dependent variable. Principal particle concentration as the independent variable Linear regression model: , and The regression coefficients can be obtained by fitting a large amount of historical production data using the least squares method, and the slope of the linear regression model is extracted. Next, the average concentrate grade within the abnormal concentrate grade time window is calculated, and the absolute difference between the average concentrate grade and the preset target concentrate grade is divided by the slope of the linear regression model to obtain the lower limit increase value of the main particle concentration benchmark. Finally, the initial main particle concentration reference range for the intermediate crushing process is obtained. , and These represent the lower and upper limits of the initial principal particle concentration baseline range, respectively. Then, This serves as the new benchmark range for principal particle concentration.

[0040] If the process is changed to the final crushing process, the goal is to reduce the tailings grade. That is, based on historical records, a system is established with tailings grade as the dependent variable. Using the proportion of effective granularity as the independent variable Linear regression model: , and For regression coefficients, and and The methods for obtaining values ​​are the same. A negative value indicates an increase in the effective particle size ratio and a decrease in tailings grade. Next, the average tailings grade within the abnormal tailings grade time window is calculated, and the difference between the average tailings grade and the preset target tailings grade is divided by the absolute value of the slope of the corresponding linear regression model to obtain the lower limit increase value of the effective particle size ratio benchmark, denoted as [missing value]. Ultimately, the initial effective particle size distribution range for the intermediate crushing process is obtained. , and These represent the lower and upper limits of the initial effective particle size ratio reference range, respectively. Then, This serves as a new benchmark range for the proportion of effective particle size.

[0041] It should be noted that the preset target concentrate grade and preset target tailings grade can both be set by production management personnel according to process requirements and manually imported.

[0042] Through the above process, the embodiments of the present invention enable the crushing process to automatically adjust the discharge standard according to the real-time working conditions, effectively improving the crushing's adaptability to changes in the properties of the incoming material, ensuring that the crushing process always provides materials with stable particle size characteristics for subsequent magnetic separation operations, improving the stability of the feed quality of the magnetic separation process, and realizing precise control of the particle size characteristics of the crushed product, thus creating favorable conditions for improving the recovery rate and concentrate grade of the subsequent magnetic separation process.

[0043] This invention, through the linkage analysis of multi-stage magnetic separation and multi-stage crushing processes, enables online assessment and change request judgment of the crushing process's operating status, shifting the control node from the magnetic separation stage to the crushing stage. This allows for intervention measures to be taken before problematic materials enter the magnetic separator. It fundamentally changes the passive situation of reactive remediation, effectively reducing the risk of metal loss due to substandard ore feed and further improving the magnetic separation control effect.

[0044] S2. Based on the discharge particle size benchmark range, combined with real-time concentrate grade and real-time tailings grade, the adjustment trigger status of each level of crushing equipment and the upward trend of tailings grade are judged in parallel.

[0045] After dynamically setting a new discharge particle size benchmark range for the changed process, a continuous monitoring and judgment phase begins. This confirms whether the actual discharge particle size of the process has indeed and continuously deviated from the new standard, thereby avoiding misoperation caused by instantaneous fluctuations and ensuring that adjustment commands are necessary and accurate. A specific judgment process example is as follows: Based on the real-time discharge particle size data collected by a particle size monitoring device installed at the discharge port of the primary jaw crusher, such as an online laser particle size analyzer, the real-time coarse particle size range is calculated.

[0046] The calculated real-time coarse particle size range is compared with the discharge particle size reference range dynamically set in step S1.

[0047] If multiple consecutive time windows of real-time coarse-grained particle size variation are detected, for example, if three consecutive time windows of continuous deviation are detected, then the primary crushing equipment is ultimately adjusted. This effectively filters out brief, sporadic particle size fluctuations.

[0048] Based on the real-time discharge particle size data of the cone crusher, the real-time particle weight percentage of all particles whose particle size falls within the preset main particle size distribution range is calculated, and this percentage is recorded as the real-time main particle concentration.

[0049] The real-time main particle concentration is compared with the dynamically set benchmark range. If the real-time main particle concentration remains below the lower limit of the benchmark range for multiple consecutive time windows, it indicates that the cone crusher is failing to stably produce a sufficient proportion of the target particle size product, requiring intervention. Ultimately, this triggers adjustments to the intermediate crushing equipment. In certain characteristic requirement scenarios, it may also be considered whether the concentration remains consistently above the upper limit.

[0050] Based on the real-time discharge particle size data of the sand making crusher, the real-time particle weight percentage of particles whose particle size falls within the particle size range that meets the requirements of the magnetic separation process is statistically calculated, and this percentage is recorded as the real-time effective particle size ratio.

[0051] The real-time effective particle size ratio is compared with the dynamically set effective particle size ratio benchmark range. If the real-time effective particle size ratio is continuously lower than the lower limit of its benchmark range for multiple consecutive time windows, it indicates that the proportion of material produced by the sand making crusher that can be directly used for efficient magnetic separation is insufficient, resulting in insufficient resource recovery. In this case, the final judgment is to trigger the adjustment of the final crushing equipment.

[0052] If each crushing process does not meet the condition of continuous deviation for multiple consecutive time windows, it is determined that the crushing equipment adjustment will not be triggered.

[0053] This invention, through continuous multi-time-window monitoring and comparison of the output particle size at each crushing stage, accurately distinguishes between temporary anomalies and persistent deviations, ensuring that equipment adjustment commands are only triggered when a problem actually exists and persists. This not only prevents oscillations in the production process due to frequent and unnecessary adjustments, ensuring the operational stability of the entire recycling process, but also effectively improves the accuracy and reliability of recycling control, effectively avoiding erroneous equipment adjustments caused by instantaneous fluctuations or detection noise.

[0054] Considering that tailings grade is one of the most critical indicators for measuring whether metal recovery is sufficient and whether the process is operating normally, an abnormal rise in tailings grade directly indicates an increase in metal loss, which must be given high priority. Therefore, this invention uses tailings grade trend judgment to show the current tailings grade situation.

[0055] In this embodiment of the invention, the linear regression slope of tailings grade is analyzed by setting at least two time windows of different lengths. The analysis comprehensively judges whether the tailings grade shows a continuous and real abnormal upward trend from different time scales. The specific implementation process for judging the upward trend of tailings grade is as follows: at least two time windows of different lengths are set, for example, a short window of 15 minutes and a long window of 60 minutes are set to analyze the tailings grade data synchronously. The short window is used to capture recent abrupt changes, and the long window is used to confirm the continuous trend, thereby taking into account both the sensitivity and stability of trend identification.

[0056] The linear regression slope of tailings grade data with respect to time was calculated for each of the above-mentioned time windows of different lengths. This slope is used to quantitatively describe the average rate of change of tailings grade at each time scale.

[0057] If the calculated linear regression slopes for all different time windows are greater than the preset growth rate, it indicates that the tailings grade shows a clear growth trend that exceeds the normal fluctuation range at different time scales. In this case, it is finally determined that the tailings grade has an abnormal upward trend.

[0058] If the linear regression slope does not exceed the set growth rate within any time window, it is determined that no abnormal upward trend has occurred. In this case, only the crushing equipment is adjusted, and no further responsibility determination or coordinated adjustment is required.

[0059] It should be noted that the set growth rate can be obtained based on statistical analysis of historical data. For example, extract the tailings grade data during the stable production period in the past year, calculate the linear regression slope of the tailings grade in multiple time windows of the same length in the past year, sort the calculated slope values ​​in ascending order, construct a slope sequence, and take the 85th percentile of the slope sequence as the current set growth rate.

[0060] S3. If the tailings grade shows an abnormal upward trend, the responsible party for the abnormality shall be determined based on the real-time feed grade of the magnetic separation process. If only the crushing equipment is adjusted, the crushing equipment shall be adjusted.

[0061] After confirming an abnormal upward trend in tailings grade, this embodiment implements an effective adjustment strategy. The specific implementation process for determining the responsible party for the anomaly through quantitative analysis of the correlation between feed grade and tailings grade includes: simultaneously collecting real-time feed grade and real-time tailings grade data sequences for the magnetic separation process within multiple identical analysis time windows. For each time window, the Pearson correlation coefficient between these two variables is calculated.

[0062] To ensure the stability of the judgment and avoid interference from randomness in a single time period, the Pearson correlation coefficients calculated over multiple consecutive time windows are arithmetically averaged to obtain the final average correlation coefficient.

[0063] If the calculated average correlation coefficient is greater than the first correlation coefficient threshold, it indicates a strong correlation between the feed grade and the tailings grade. That is, as the feed grade increases, the tailings grade also increases synchronously, reflecting the current crushing-magnetic separation process's insufficient adaptability to high-grade feeds, and potentially inherent process problems such as poor particle size and decreased separation efficiency.

[0064] If the average correlation coefficient is lower than the second correlation coefficient threshold, it indicates a weak correlation between the feed grade and the tailings grade. This suggests that the abnormal rise in tailings grade is not closely related to changes in the grade of upstream feed material, and may be due to a significant deviation of the crushed particle size from the standard, a malfunction in the magnetic separation equipment itself (such as insufficient magnetic field strength), or other external operational factors.

[0065] When either of the above two situations occurs, the responsible party is determined to be another party, which means that coordinated adjustments to the crushing equipment and magnetic separation equipment are required.

[0066] If the average correlation coefficient is between two thresholds, it indicates a moderate correlation between the feed grade and the tailings grade. This is a typical response during normal production. In this case, the responsible party is determined to be the fluctuation in the feed grade, and the crushing equipment should be adjusted accordingly.

[0067] Understandably, in one specific embodiment, the first correlation coefficient threshold and the second correlation coefficient threshold are set based on the following rules: First, at least 6 months of historical production data are collected, the Pearson correlation coefficient distribution of ore grade and tailings grade is statistically analyzed, and sorted from smallest to largest. The 25th percentile of the historical correlation coefficient distribution is taken as the first correlation coefficient threshold, and the 75th percentile of the historical correlation coefficient distribution is taken as the second correlation coefficient threshold.

[0068] This invention, through its liability determination rule, can clearly distinguish whether a problem originates from changes in upstream materials or from a failure in its own process. This provides a key decision-making basis for taking the most accurate and effective adjustment measures, avoids ineffective equipment parameter adjustments, and effectively improves the overall control efficiency in complex operating conditions.

[0069] Considering that when only the crushing equipment adjustment is triggered, the final tailings condition meets the final requirements, only the settings of the crushing equipment, such as the feed rate, need to be adjusted.

[0070] Based on this, in a specific embodiment, the specific implementation process for adjusting the crushing equipment includes: if the adjustment is triggered in the primary crushing equipment, it indicates that the main problem is unstable feeding or an excessively large range of coarse particles in the output. The feed rate of the jaw crusher is then reduced by a preset level. For example, if the current feed rate is level 4 (corresponding to 100 tons / hour), it is reduced to level 3 (corresponding to 80 tons / hour). Furthermore, by reducing the feed rate, the risk of blockage in the crushing chamber is mitigated, and the residence time of the material in the chamber is extended, thereby achieving a more thorough and uniform crushing effect. This helps to reduce the range of coarse particles in the output and stabilize the feeding of subsequent processes.

[0071] If the adjustment is triggered for the intermediate crushing equipment, it indicates that the main particle concentration of its output does not meet the requirements. In this case, first determine the direction of deviation of the current real-time main particle concentration from its reference range. If the current main particle concentration is greater than the upper limit of its output particle size reference range, it indicates that the product is too fine. In this case, appropriate action can be taken, usually maintaining the crushing equipment settings. If the current main particle concentration is lower than the lower limit of its reference range, it indicates that the set target particle proportion is insufficient. In this case, perform one or a combination of the following adjustments: increase the current spindle speed of the cone crusher to the next higher rotational speed level, such as from 600 rpm to 650 rpm.

[0072] The discharge orifice diameter of the cone crusher is reduced to a smaller size (e.g., from 50mm to 45mm). Increasing the rotational speed increases the crushing frequency, while reducing the discharge orifice diameter directly controls the maximum output particle size. The synergistic effect of both effectively improves the concentration of the target particle size in the product. In one specific embodiment, one or two combinations can be selected based on the deviation of the main particle concentration from its lower limit of the reference range. When the deviation is small, one method is typically chosen; when the deviation is large, two combinations can be selected. The specific selection method can be determined by the implementer. In this invention, when the deviation is small, it is preferable to reduce the discharge orifice diameter of the cone crusher to a smaller size.

[0073] If the final crushing equipment adjustment is triggered, it indicates that the effective particle size ratio of its output is insufficient, which is the key to metal loss. At this time, the current spindle speed of the sand making crusher should be increased to the next rotation speed level, and the current discharge port diameter of the equipment should be reduced to the previous diameter level.

[0074] Understandably, for final crushing, more aggressive measures are needed to increase the yield of fine-grained products. Simultaneously increasing the rotational speed and reducing the discharge opening can effectively enhance crushing intensity, working together to increase the proportion of effective particles, thus creating optimal conditions for magnetic separation and recovery.

[0075] After adjusting any of the above crushing equipment, the effect verification stage begins: continuously collect the discharge particle size data of the adjusted equipment in real time, and calculate the corresponding coarse particle size span, main particle concentration and effective particle size ratio.

[0076] If the adjusted discharge particle size data meets its discharge particle size reference range, the adjustment is deemed effective, and the current adjustment parameters of the crushing equipment are maintained and operation continues.

[0077] If, after a reasonable stabilization period, the output particle size data still fails to meet the baseline range, it indicates that the equipment parameter adjustment has failed to resolve the issue, and there is a high probability of equipment malfunction or other unforeseen process failures. In this case, an equipment malfunction warning command should be initiated and sent to the control room's operating interface or the maintenance personnel's mobile terminal, prompting manual intervention for inspection and repair to prevent continued production losses due to latent equipment failures.

[0078] S4. If the abnormality is caused by fluctuations in ore grade, then the crushing equipment will be adjusted. If the abnormality is caused by other factors, a collaborative adjustment strategy will be generated and executed.

[0079] When the responsible party is determined to be the feed grade fluctuation, it indicates that there is a clear and expected transmission relationship between the tailings anomaly and the upstream material change. At this time, the main problem is that the crushing equipment has failed to adapt to the change in feed conditions and its output particle size has failed to meet the requirements. Therefore, only the crushing equipment needs to be adjusted.

[0080] When the responsible party is determined to be "other," it indicates that the abnormal tailings are not correlated with the fluctuations in the feed. The root cause may be a decrease in the efficiency of the magnetic separation equipment itself, improper magnetic field strength parameters, or other process defects besides the crushing particle size. If only the crushing equipment is adjusted at this time, it will be difficult to address the root cause of the problem and will not be able to effectively curb metal loss.

[0081] Based on this, the specific implementation process of generating and executing the coordinated adjustment strategy includes: calculating the relative deviation between the linear regression slope and the set growth rate based on the linear regression slope of tailings grade within different time window lengths, which serves as the degree of deviation. The larger the degree of deviation, the faster the abnormal rate at that time scale. Furthermore, for ease of subsequent analysis, the boundary value for the degree of deviation is set to 1.

[0082] The baseline anomaly index is obtained by linearly weighting and summing the deviations across different time windows. Subsequently, the ratio of the number of time windows in which the linear regression slope is greater than the set growth rate to the total number of time windows is calculated, and this ratio is used as the excess correction factor. This factor reflects the prevalence of anomalous trends across different time scales. It further adjusts the baseline index using an excess correction factor to output the final anomaly index. , Furthermore, this correction ensures that an anomalous trend that is prevalent across all time windows will be assessed with a higher severity level, making it more targeted.

[0083] The preset adjustment strategy knowledge base stores the mapping relationship between anomaly levels and the set of adjustable process parameters and their reference adjustment ranges.

[0084] The anomaly level is based on the anomaly severity index. The numerical range is defined.

[0085] The set of adjustable process parameters includes at least one parameter selected from the group consisting of the discharge particle size reference range of each crushing stage, the spindle speed of each crushing equipment, the discharge port diameter, and the magnetic field strength of each magnetic separator.

[0086] The reference adjustment range is a pre-set parameter adjustment amount or direction corresponding to the abnormality level.

[0087] The adjustment strategy knowledge base is constructed through the following steps: First, data collection is carried out, collecting historical abnormal cases in the high-carbon ferrochrome tailings crushing and recycling process. Each case includes an abnormality index, the process parameters adjusted during the abnormality duration and their adjustment range, and the evaluation of the control effect after adjustment.

[0088] Then, the cases are categorized, and the historical abnormal cases are classified into different abnormal levels according to the abnormality index.

[0089] Next, strategy extraction is performed. For each anomaly level, the frequency of the adjusted process parameters in all historical anomaly cases under that level is statistically analyzed. Parameters with a frequency higher than a preset threshold are summarized into the set of adjustable process parameters corresponding to that level. Based on the statistical distribution of the adjustment range, the reference adjustment range for each parameter is determined.

[0090] Finally, a knowledge base is generated, storing the mapping relationship between the anomaly level, the set of adjustable process parameters, and the reference adjustment range to form the adjustment strategy knowledge base. As a preferred example, when the anomaly severity index represents a mild anomaly, i.e. The set of adjustable process parameters is: {lower limit of effective particle size ratio reference range, magnetic field strength of secondary magnetic separation, and main shaft speed of cone crusher}.

[0091] When the anomaly severity index represents moderate anomaly, i.e. The set of adjustable process parameters is as follows: {lower limit of main particle concentration, lower limit of effective particle size ratio, magnetic field strength of primary magnetic separation, and spindle speed of secondary cone crusher}.

[0092] When the anomaly severity index represents severe anomaly, i.e. The set of adjustable process parameters is {the reference range of each crushing stage, the magnetic field strength of each magnetic separation stage, and the spindle speed of each crushing equipment}.

[0093] It should be noted that the range of values ​​for the abnormality index corresponding to the different degrees of abnormality mentioned above can be adjusted according to the actual recovery requirements. The values ​​given are only a preferred example of the present invention and are not fixed values.

[0094] For each adjustable process parameter in the matched parameter set, a preset adjustment range is matched for the parameter based on the current anomaly index and its level, and the adjustment is automatically executed according to the matched range. The adjustment direction is based on positive optimization, such as raising the lower limit of the discharge particle size, enhancing the magnetic field strength, and increasing the spindle speed to improve the fine crushing effect.

[0095] After all adjustments are implemented, the trend of tailings grade changes is continuously monitored. If the abnormal upward trend does not ease, the current anomaly level is automatically upgraded by one level, such as from moderate to severe anomaly, and the adjustment parameter set and adjustment range corresponding to the next level are retrieved. If the abnormal trend of tailings grade continues to worsen after multiple upgrades and adjustments, it is determined that the current strategy is no longer effective, and a manual intervention warning command is initiated.

[0096] This invention achieves precise control of particle size parameters in the current process by adjusting the crushing equipment. At the same time, through the coordinated adjustment of the crushing and magnetic separation equipment, it not only ensures that the output quality of each crushing stage meets the magnetic separation requirements, but also further optimizes the control effect through the linkage adjustment of multiple equipment parameters under complex working conditions.

[0097] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for controlling the multi-stage crushing, recovery, and magnetic separation of high-carbon ferrochrome tailings, characterized in that, The method includes: Based on historical discharge records, real-time feed flow rate of each level of magnetic separation process, and real-time total recovery rate, the discharge particle size reference range of each level of crushing process is dynamically set. Before setting the discharge particle size reference range, a change determination process is also included. The specific determination process includes: Calculate the standard deviation of feed flow rate and the standard deviation of total recovery rate within the current time window; If the standard deviation of the feed flow rate or total recovery rate of the magnetic separation process exceeds the historical stable range, and the standard deviation of the output coarse particle size range of the primary crushing process also exceeds its historical stable range, then the primary crushing process will be set as a change process. Based on a preset fixed time interval, the real-time concentrate grade and real-time tailings grade are divided into multiple continuous analysis time windows. When the concentrate grade continuously declines and falls below the historical average, or when the tailings grade continuously rises and rises above the historical average, the analysis time window of continuous decline or continuous rise is considered an abnormal time window. Within the abnormal time window, calculate the Pearson correlation coefficients between the concentrate grade and tailings grade and the concentration of main particles in the intermediate crushing output and the proportion of effective particle sizes in the final crushing output, respectively, and set up change processes based on the magnitude of the correlation coefficients. The process of dynamically setting the discharge particle size reference range includes: If the changed process is a primary crushing process, select historical records that are within the preset critical time window from the historical records and record them as candidate historical records; Based on the standard deviation of feed flow rate and the standard deviation of total recovery rate, the coefficient of variation of feed flow rate and the coefficient of variation of total recovery rate are calculated, and the weighted sum of the two is used as the load fluctuation index. Select the historical record with the smallest load fluctuation index from the candidate historical records, and update the primary crushing coarse particle size range corresponding to the historical record to the discharge particle size reference range of the primary crushing process. If the process is changed to intermediate crushing, a linear regression model of concentrate grade and main particle concentration is established based on historical records, and the slope of the linear regression model is extracted. Calculate the average concentrate grade within the abnormal concentrate grade time window, calculate the difference between the average concentrate grade and the preset target concentrate grade, and record the ratio of the difference to the slope as the lower limit increase value of the main particle concentration benchmark. The sum of the lower limit of the initial main particle concentration benchmark range and the benchmark lower limit increase value of the intermediate crushing process is updated to the new lower limit of the main particle concentration benchmark range, while its upper limit remains unchanged. If the process is changed to the final crushing process, a linear regression model of tailings grade and effective particle size ratio is established based on historical records, and the effective particle size ratio benchmark range is updated according to the update method of the main particle concentration benchmark range. Based on the aforementioned discharge particle size reference range, combined with real-time concentrate grade and real-time tailings grade, the adjustment trigger status of each stage of crushing equipment and the upward trend of tailings grade are determined in parallel. If the tailings grade shows an abnormal upward trend, the responsible party will be determined based on the real-time feed grade of the magnetic separation process; if it only triggers the adjustment of the crushing equipment, then the crushing equipment will be adjusted. If the abnormality is caused by fluctuations in ore grade, the crushing equipment will be adjusted; if the abnormality is caused by other factors, a collaborative adjustment strategy will be generated and executed.

2. The method for multi-stage crushing, recycling, and magnetic separation control of high-carbon ferrochrome tailings as described in claim 1, characterized in that: For non-change processes, the initial discharge particle size reference range shall be used as the final discharge particle size reference range.

3. The method for multi-stage crushing, recycling, and magnetic separation control of high-carbon ferrochrome tailings as described in claim 1, characterized in that: The determination of the adjustment trigger status of each stage of crushing equipment includes: Based on the real-time discharge particle size data of the primary crushing process, the real-time coarse particle size span is calculated. If the coarse particle size range deviates from the discharge particle size reference range for multiple consecutive time windows, it is determined that the primary crushing equipment adjustment is triggered. Based on the real-time discharge particle size data of the intermediate crushing process, the real-time particle weight percentage of particles whose particle size is within the preset main particle size distribution range is counted and recorded as the real-time main particle concentration. If the concentration of main particles is consistently lower than the lower limit of the corresponding discharge particle size reference range for multiple consecutive time windows, it is determined that the adjustment of the intermediate crushing equipment is triggered. Based on the real-time discharge particle size data of the final crushing process, the percentage of particle weight within the particle size range required by the magnetic separation process is statistically calculated in real time and recorded as the real-time effective particle size ratio. If the effective particle size percentage is consistently lower than the lower limit of the corresponding discharge particle size reference range for multiple consecutive time windows, it is determined that the final crushing equipment adjustment is triggered. Otherwise, the crushing equipment adjustment will not be triggered.

4. The method for multi-stage crushing, recycling, and magnetic separation control of high-carbon ferrochrome tailings as described in claim 1, characterized in that: The judgment of the upward trend in tailings grade includes: Set at least two different time window lengths and calculate the linear regression slope of tailings grade within different time window lengths; If the linear regression slope of tailings grade is greater than the set growth rate within different time windows, it is judged that the tailings grade has an abnormal upward trend; otherwise, it is judged that no abnormal upward trend has occurred.

5. The method for multi-stage crushing, recycling, and magnetic separation control of high-carbon ferrochrome tailings as described in claim 1, characterized in that: The entities responsible for identifying abnormalities include: Calculate the Pearson correlation coefficient between feed grade and tailings grade within the same time window; The average correlation coefficient was obtained by averaging the Pearson correlation coefficients between ore grade and tailings grade over multiple time windows. If the average correlation coefficient indicates a strong or weak correlation, the responsible party is others; if the average correlation coefficient indicates a correlation between a weak or strong correlation, the responsible party is the fluctuation in ore grade.

6. The method for multi-stage crushing, recycling, and magnetic separation control of high-carbon ferrochrome tailings as described in claim 1, characterized in that: The adjustment of the crushing equipment includes: If the primary crushing equipment adjustment is triggered, the feed rate of the primary crushing equipment will be reduced to the previous preset feed rate level; If the adjustment of the intermediate crushing equipment is triggered, if the current main particle concentration is less than the lower limit of its discharge particle size reference range, the corresponding current spindle speed of the intermediate crushing equipment will be increased to the next rotation speed level, and / or its discharge port diameter will be reduced to the previous diameter level. If the final crushing equipment adjustment is triggered, the current spindle speed of the final crushing equipment will be increased to the next rotational speed level, and the current discharge port diameter will be reduced to the previous diameter level. After the crushing equipment adjustment is completed, the discharge particle size data of the corresponding crushing equipment is continuously collected in real time. If the current discharge particle size meets the benchmark range, the current adjustment parameters of the crushing equipment are maintained; otherwise, an equipment abnormality warning command is initiated and sent.

7. The method for multi-stage crushing, recycling, and magnetic separation control of high-carbon ferrochrome tailings as described in claim 4, characterized in that: The generation and execution of the collaborative adjustment strategy includes: Based on the linear regression slope of tailings grade within different time windows, an anomaly index is calculated, and the set of adjustable process parameters and corresponding adjustment ranges corresponding to the anomaly level of the anomaly index are matched from the preset adjustment strategy knowledge base. For each adjustable process parameter, perform the corresponding adjustment according to the stated adjustment range; After all adjustments are implemented, the trend of tailings grade changes will be continuously monitored. If the abnormal upward trend does not ease, the preset adjustment range of the next level below the current abnormal level will be retrieved for an upgrade adjustment. If the situation continues to worsen after multiple adjustments, a manual intervention warning will be initiated. When the abnormal trend eases, maintain the current strategy until the tailings grade no longer shows an abnormal upward trend.

8. The method for multi-stage crushing, recycling, and magnetic separation control of high-carbon ferrochrome tailings as described in claim 7, characterized in that: The calculation of the anomaly index includes: Calculate the relative deviation between the linear regression slope and the set growth rate to obtain the degree of deviation for different time windows; The ratio of the number of time windows in which the linear regression slope is greater than the set growth rate to the total number of time windows is used as the excess correction factor. The weighted sum of the deviations in different time windows is used as the baseline anomaly index. The baseline anomaly index is then corrected based on the excess correction factor to output the final anomaly index.

Images