A method and system for controlling a composite mineral admixture ball mill

Abstract

The application relates to the technical field of ball mill control, and discloses a composite mineral admixture ball mill control method and system, which comprises the following steps: collecting driving current data, mill sound data and vibration data, and constructing current sequences, mill sound sequences and vibration sequences of a collection period; marking a target collection period, determining a mill sound first slope, a mill sound wear confidence, a current first slope, a current wear confidence, a vibration first slope and a vibration wear confidence of the target collection period, and determining a steel ball wear coefficient of the target collection period; determining a wear difference coefficient of the target collection period; and controlling the rotating speed of the ball mill of the composite mineral admixture according to the wear difference coefficient of the target collection period and the rotating speed. The application can realize self-adaptive control of the rotating speed of the ball mill of the composite mineral admixture.

Description

Technical Field

[0001] This application relates to the field of ball mill control technology, specifically to a control method and system for a ball mill containing composite mineral admixtures. Background Technology

[0002] Ball mills, after ball addition, finely grind ultrafine composite mineral admixtures, achieving ultrafine and uniform particle size, enhancing the activity and reactivity of the mineral admixtures, optimizing particle morphology and gradation, and promoting the resource utilization of solid waste. Precise control of the ball-to-material ratio and rotational speed during the fine grinding process in ball mills can improve production efficiency and ensure the stability of product quality.

[0003] When grinding ultrafine composite mineral admixtures in a ball mill, the mineral blending ratio is fixed, so the rotation speed and ball-to-material ratio are generally set as fixed parameters. However, during the grinding process, the steel balls are continuously worn down, and the ball-to-material ratio inside the ball mill also fluctuates dynamically with the wear of the steel balls. If a fixed rotation speed is still used at this time, it will affect the grinding quality and efficiency of the ball mill, and affect the stability of product quality. Summary of the Invention

[0004] This application provides a control method and system for a ball mill containing composite mineral admixtures to solve the problem of mismatch between a fixed rotational speed and the gradual wear of steel balls during the fine grinding process in the ball mill, which affects the quality of the ground products. The specific technical solution adopted is as follows:

[0005] In a first aspect, one embodiment of this application provides a method for controlling a ball mill containing composite mineral admixtures, the method comprising the following steps:

[0006] Drive current data, grinding sound data, and vibration data are collected at different times within different acquisition cycles to construct the current sequence, grinding sound sequence, and vibration sequence for each acquisition cycle.

[0007] Any acquisition period is designated as the target acquisition period. Based on the changing trend of the grinding sound data within the grinding sound sequence of the target acquisition period, and the difference between the grinding sound sequence of the target acquisition period and the previous adjacent acquisition period, the first slope of the grinding sound and the grinding sound wear confidence level of the target acquisition period are determined. Following the method of determining the first slope of the grinding sound and the grinding sound wear confidence level of the target acquisition period based on the grinding sound sequence of the target acquisition period, the first slope of the current, the current wear confidence level, the first slope of the vibration, and the vibration wear confidence level of the target acquisition period are determined based on the current sequence and the vibration sequence of the target acquisition period, respectively. Combining the first slope of the grinding sound and the grinding sound wear confidence level, and the changing trend of the vibration data within the vibration sequence of the target acquisition period, the steel ball wear coefficient of the target acquisition period is determined.

[0008] The wear difference coefficient for the target acquisition cycle is determined based on the differences in current sequence, grinding sound sequence, and vibration sequence between the target acquisition cycle and the first acquisition cycle, as well as the difference in steel ball wear coefficient between the target acquisition cycle and the first acquisition cycle.

[0009] Based on the wear difference coefficient and rotation rate of the target collection cycle, the rotation rate of the ball mill for composite mineral admixtures is controlled.

[0010] Furthermore, the method for determining the first slope of the grinding sound is as follows:

[0011] A linear fit is performed on the grinding sound sequence of the target acquisition period, and the slope of the fitted line is denoted as the first slope of the grinding sound of the target acquisition period.

[0012] Furthermore, the method for determining the confidence level of the grinding sound wear is as follows:

[0013] The first slope of the grinding sound in the target acquisition cycle is taken as the exponent of the exponential function with the natural constant as the base, and the reciprocal of the calculated value of the exponential function is recorded as the first exponential value of the grinding sound in the target acquisition cycle.

[0014] The first ratio of the grinding sound in the target acquisition period is determined based on the difference between the grinding sound sequence of the target acquisition period and the previous adjacent acquisition period.

[0015] The positive correlation between the first ratio of grinding sound and the first index value of grinding sound in the target acquisition period is recorded as the grinding sound wear confidence level of the target acquisition period.

[0016] Furthermore, the method for determining the first ratio of the grinding sound in the target acquisition cycle is as follows:

[0017] The ratio of the previous adjacent acquisition period of the target acquisition period to the mean value of the grinding sound data contained in the grinding sound sequence of the target acquisition period is denoted as the first grinding sound ratio of the target acquisition period.

[0018] Furthermore, the method for determining the steel ball wear coefficient for the target acquisition period by combining the first slope of the grinding sound and the confidence level of the grinding sound, as well as the changing trend of vibration data within the vibration sequence of the target acquisition period, includes the following specific methods:

[0019] The number of negative values ​​among the first slope of grinding sound, the first slope of current, and the first slope of vibration in the target acquisition cycle is recorded as the number of descents in the target acquisition cycle.

[0020] The average of the confidence scores for noise wear, current wear, and vibration wear during the target acquisition cycle is denoted as the wear confidence score for the target acquisition cycle.

[0021] The vibration sequence of the target acquisition period is divided into a first preset number of vibration subsequences. The kurtosis of each vibration subsequence is calculated. Based on the changing trend of the kurtosis of all vibration subsequences divided from the vibration sequence of the target acquisition period, the kurtosis ratio of the target acquisition period is determined.

[0022] The positive correlation between the number of drops, wear confidence, and kurtosis ratio in the target acquisition cycle is recorded as the steel ball wear coefficient of the target acquisition cycle.

[0023] Furthermore, the method for determining the kurtosis ratio of the target acquisition period is as follows:

[0024] The normalized value of the slope of the fitted line obtained by fitting the kurtosis of all vibration subsequences divided from the vibration sequence of the target acquisition period is denoted as the kurtosis slope of the target acquisition period.

[0025] The range of kurtosis of all vibration subsequences divided from the vibration sequence of the target acquisition period is denoted as the kurtosis range of the target acquisition period.

[0026] The ratio of the kurtosis range to the kurtosis slope of the target acquisition period is denoted as the kurtosis ratio of the target acquisition period.

[0027] Furthermore, the method for determining the wear difference coefficient of the target acquisition cycle is as follows:

[0028] The difference between the steel ball wear coefficient of the target acquisition cycle and the first acquisition cycle is denoted as the steel ball wear difference of the target acquisition cycle.

[0029] The initial trend similarity of the target acquisition cycle is determined based on the differences in current sequence, grinding sound sequence, and vibration sequence between the target acquisition cycle and the first acquisition cycle.

[0030] The normalized value of the ratio of the steel ball wear difference in the target acquisition cycle to the initial trend similarity is denoted as the wear difference coefficient of the target acquisition cycle.

[0031] Furthermore, the method for determining the initial trend similarity of the target acquisition period is as follows:

[0032] Sequence decomposition is performed on the grinding sound sequence of the target acquisition period to obtain the trend sequence of the grinding sound sequence of the target acquisition period. The absolute value of the similarity between the trend sequence of the grinding sound sequence of the target acquisition period and the first acquisition period is denoted as the initial trend similarity of the grinding sound of the target acquisition period.

[0033] The vibration sequence of the target acquisition period is decomposed to obtain the trend sequence of the vibration sequence of the target acquisition period; the absolute value of the similarity between the trend sequence of the vibration sequence of the target acquisition period and the first acquisition period is denoted as the initial trend similarity of the vibration of the target acquisition period.

[0034] The current sequence of the target acquisition period is decomposed to obtain the trend sequence of the current sequence of the target acquisition period; the absolute value of the similarity between the trend sequence of the current sequence of the target acquisition period and the current sequence of the first acquisition period is denoted as the initial trend similarity of the current of the target acquisition period.

[0035] The average of the initial trend similarity of the grinding sound, the initial trend similarity of the vibration, and the initial trend similarity of the current in the target acquisition cycle is denoted as the initial trend similarity of the target acquisition cycle.

[0036] Furthermore, the specific method for controlling the rotational speed of the ball mill for composite mineral admixtures based on the wear difference coefficient and rotational speed during the target acquisition cycle includes:

[0037] The product of the wear difference coefficient of the target acquisition cycle and the preset speed adjustment value is recorded as the adjustment amount of the target acquisition cycle. The sum of the adjustment amount of the target acquisition cycle and the speed is recorded as the adjustment speed of the next adjacent acquisition cycle of the target acquisition cycle.

[0038] The rotational speed of the target acquisition cycle is used as the rotational speed value of the next adjacent acquisition cycle.

[0039] Secondly, embodiments of this application also provide a control system for a ball mill containing composite mineral admixtures, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0040] The beneficial effects of this application are:

[0041] This application considers that wear of the steel balls leads to a gradual decrease in their volume and mass. At the same preset rotational speed, smaller, lighter steel balls generate weaker impact forces. The collected drive current, grinding sound, and vibration data all show a slow and gradual decreasing trend. Based on the changing trends of these data, the degree of wear on the steel balls within the corresponding acquisition period is evaluated, and the steel ball wear coefficient for the target acquisition period is determined. A higher steel ball wear coefficient indicates a greater degree of wear on the steel balls within the corresponding acquisition period. In this case, the wear on the steel balls is less compatible with the preset rotational speed, and the impact of steel ball wear on the grinding efficiency of the ball mill is greater. Therefore, the corresponding rotational speed should be increased to improve the grinding efficiency of the ball mill. The grinding efficiency of the ball mill is improved. Furthermore, based on the difference in wear degree of the steel balls between each collection cycle and the first collection cycle, the error in evaluating the wear condition of the steel balls is reduced. The difference between the movement state of the steel balls within the ball mill during the corresponding collection cycle and the movement state at the initial grinding stage is evaluated, and the wear difference coefficient for the target collection cycle is determined. Finally, based on the wear difference coefficient and rotational speed of the target collection cycle, the rotational speed of the ball mill for composite mineral admixtures is controlled. This solves the problem of mismatch between a fixed rotational speed and the gradual wear of steel balls during the fine grinding process, which affects the quality of the ground product. Real-time adjustment of the rotational speed during the fine grinding process of the ball mill improves the grinding control effect of the ball mill, increases grinding efficiency, and reduces the energy consumption of the ball mill's energy-saving motor. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a schematic flowchart illustrating a ball mill control method for composite mineral admixtures provided in one embodiment of this application.

[0044] Figure 2 This is a flowchart illustrating the process of obtaining the confidence level of grinding sound wear according to one embodiment of this application. Detailed Implementation

[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] Please see Figure 1 The diagram illustrates a flowchart of a ball mill control method for composite mineral admixtures according to an embodiment of this application. The method includes the following steps:

[0047] Step S001: Collect driving current data, grinding sound data and vibration data at different acquisition times within different acquisition cycles, and construct the current sequence, grinding sound sequence and vibration sequence for each acquisition cycle.

[0048] A current sensor is installed on the energy-saving motor of the ball mill, and a grinding noise meter is installed on the ground below the ball mill cylinder, with the grinding noise meter angled directly towards the ball mill. A vibration acceleration sensor is installed on the bearing housing of the ball mill. During the period from adding balls to the ball mill until the next ball addition, while the ball mill is grinding ultrafine composite mineral admixtures, the current sensor collects the current data of the energy-saving motor, the grinding noise meter collects the grinding noise data, and the vibration acceleration sensor collects the vibration data.

[0049] In this embodiment, the current data acquisition frequency is 30Hz, and the average value of all current data acquired within one second is recorded as the driving current data for the corresponding second. In this embodiment, the acquisition frequency of grinding sound data and vibration data is once per second, and a total of 10 minutes of driving current data, grinding sound data and vibration data are acquired in one acquisition cycle.

[0050] The driving current data, grinding sound data, and vibration data collected within the same acquisition period are arranged in the order of acquisition to form the current sequence, grinding sound sequence, and vibration sequence of the same acquisition period.

[0051] It should be noted that, for ease of calculation, all drive current data, grinding sound data, and vibration data involved in the calculation in this embodiment have undergone data preprocessing to eliminate the influence of dimensions. This embodiment uses the Z-Score standard normalization method to perform dimensionless processing on the drive current data, grinding sound data, and vibration data respectively. In practical applications, implementers can use other methods such as the existing technology of maximum-minimum value normalization for dimensionless processing, which are not limited here.

[0052] Thus, the current sequence, grinding sound sequence, and vibration sequence for each acquisition cycle are obtained.

[0053] Step S002: Record any acquisition cycle as the target acquisition cycle. Based on the changing trend of the grinding sound data within the grinding sound sequence of the target acquisition cycle, and the difference between the grinding sound sequences of the target acquisition cycle and the previous adjacent acquisition cycle, determine the first grinding sound slope and grinding sound wear confidence level of the target acquisition cycle. Following the method of determining the first grinding sound slope and grinding sound wear confidence level of the target acquisition cycle based on the grinding sound sequence, determine the first current slope, current wear confidence level, first vibration slope, and vibration wear confidence level of the target acquisition cycle based on the current sequence and vibration sequence, respectively. Combining the first grinding sound slope and grinding sound wear confidence level, and the changing trend of the vibration data within the vibration sequence of the target acquisition cycle, determine the steel ball wear coefficient of the target acquisition cycle.

[0054] When a ball mill driven by an energy-saving motor begins grinding, the proportions of each material, the ball-to-material ratio, and the rotation speed in the ultrafine composite mineral admixture are relatively stable. Therefore, the collected drive current, grinding sound, and vibration data are also relatively stable. As the grinding process progresses, collisions occur between the steel balls and the admixture, between steel balls themselves, and between the steel balls and the mill cylinder. This leads to gradual wear of the steel balls and a decrease in their impact force. At this point, the fixed rotation speed is mismatched with the gradual wear of the steel balls during the fine grinding process. If the initially set fixed rotation speed is still used, the grinding quality and efficiency will decrease. Therefore, it is necessary to analyze the wear degree of the steel balls based on the drive current, grinding sound, and vibration data collected in each sampling cycle from the time the balls are added to the next ball addition. This allows for adjustment of the rotation speed to improve grinding efficiency and save energy.

[0055] Specifically, wear of the steel ball causes its volume and mass to gradually decrease. At the same preset rotational speed, the smaller and lighter steel ball can generate a weaker impact force, and the collected grinding sound data shows a slow and gradually decreasing trend.

[0056] Let any one acquisition period be denoted as the target acquisition period.

[0057] A linear fit is performed on the grinding sound sequence of the target acquisition period to obtain the fitted line. The slope of the fitted line is recorded as the first slope of the grinding sound of the target acquisition period.

[0058] The first slope of the grinding sound is used to evaluate the changing trend of the grinding sound data collected within the corresponding acquisition period. When the first slope of the grinding sound is negative and smaller, the decreasing trend of the grinding sound data collected within the corresponding acquisition period is more obvious, and the steel ball is more likely to have worn out within the corresponding acquisition period, and the wear is more severe.

[0059] This embodiment uses a polynomial fitting algorithm to achieve linear fitting of the grinding sound sequence. Using a polynomial fitting algorithm to achieve linear fitting is a well-known technique and will not be elaborated further. In practical applications, as other implementation methods, while achieving the goal of linear fitting, implementers can use other existing methods such as the least squares method or the population mean method to fit the line; this application does not impose any special limitations.

[0060] The ratio of the previous adjacent acquisition period of the target acquisition period to the mean value of the grinding sound data contained in the grinding sound sequence of the target acquisition period is denoted as the first grinding sound ratio of the target acquisition period.

[0061] The first grinding sound ratio is used to evaluate the degree of change of the grinding sound data collected in the corresponding acquisition period compared with the average grinding sound data collected in the previous adjacent acquisition period. When the first grinding sound ratio is larger, the degree of change of the grinding sound data collected in the corresponding acquisition period is greater than the average grinding sound data collected in the previous adjacent acquisition period. The steel ball is more likely to wear in the acquisition period corresponding to the first grinding sound ratio, and the wear is more severe.

[0062] It should be noted that for the first acquisition cycle, there is no corresponding previous adjacent acquisition cycle. The first ratio of the grinding sound in the first acquisition cycle is directly assigned a value of 1.

[0063] The first slope of the grinding sound in the target acquisition period is taken as the exponent of an exponential function with the natural constant as the base. The reciprocal of the calculated value of the exponential function is denoted as the first exponential value of the grinding sound in the target acquisition period. The positive correlation between the first ratio of the grinding sound in the target acquisition period and the first exponential value of the grinding sound is denoted as the grinding sound wear confidence level in the target acquisition period.

[0064] It is understood that a positive correlation is applied to the first ratio and the first index of the grinding sound, ensuring that the first ratio and the first index of the grinding sound are positively correlated with the wear confidence level. It is understood that the positive correlation in this application refers to the relationship between the independent and dependent variables, where the independent variables are the first ratio and the first index of the grinding sound, and the dependent variable is the wear confidence level. A positive correlation means that the dependent variable increases (decreases) as the independent variable increases (decreases), and can be an additive or multiplicative relationship.

[0065] Preferably, as an embodiment of this application, the product of the first ratio of the grinding sound in the target acquisition period and the first index value is recorded as the grinding sound wear confidence level of the target acquisition period.

[0066] In practical applications, as another implementation method, the sum of the first ratio of the grinding sound and the first index value of the target acquisition cycle is recorded as the grinding sound wear confidence level of the target acquisition cycle.

[0067] The grinding sound wear confidence score is used to evaluate the probability that the steel ball has undergone wear, as reflected in the grinding sound data collected within a corresponding acquisition period. A higher grinding sound wear confidence score indicates a greater probability of wear occurring within the corresponding acquisition period, and more severe wear is expected. The flowchart for obtaining the grinding sound wear confidence score is shown below. Figure 2 As shown.

[0068] Because the initial size of the ultrafine composite mineral admixture is small, as the steel balls wear down, the amount of steel balls and ore carried by the ball mill cylinder decreases under the drive of the energy-saving motor, reducing the load on the ball mill. Consequently, the current output by the energy-saving motor in the ball mill also gradually decreases. Simultaneously, as the steel balls wear down, the intensity of the vibration generated by the steel balls on the cylinder also weakens.

[0069] According to the method of determining the first slope and wear confidence of the grinding sound in the target acquisition period based on the grinding sound sequence of the target acquisition period, the first slope and wear confidence of the current in the target acquisition period are obtained based on the current sequence of the target acquisition period, and the first slope and wear confidence of the vibration in the target acquisition period are obtained based on the vibration sequence of the target acquisition period.

[0070] The specific methods for obtaining the first current slope and current wear confidence level for the target acquisition period are as follows: A straight line is fitted to the current sequence of the target acquisition period to obtain the fitted straight line. The slope of the fitted straight line is recorded as the first current slope of the target acquisition period. The ratio of the current data in the previous adjacent acquisition period to the mean of the current data contained in the current sequence of the target acquisition period is recorded as the first current ratio of the target acquisition period. The first current slope of the target acquisition period is taken as the exponent of an exponential function with the natural constant as the base. The reciprocal of the calculated value of the exponential function is recorded as the first current exponent value of the target acquisition period. The positive correlation result between the first current ratio and the first current exponent value of the target acquisition period is recorded as the current wear confidence level of the target acquisition period.

[0071] The specific methods for obtaining the first vibration slope and vibration wear confidence level for the target acquisition period are as follows: A linear fit is performed on the vibration sequence of the target acquisition period to obtain the fitted line. The slope of the fitted line is recorded as the first vibration slope of the target acquisition period. The ratio of the mean of the vibration data contained in the vibration sequence of the target acquisition period to the mean of the previous adjacent acquisition period is recorded as the first vibration ratio of the target acquisition period. The first vibration slope of the target acquisition period is used as the exponent of an exponential function with the natural constant as the base. The reciprocal of the calculated value of the exponential function is recorded as the first vibration exponent value of the target acquisition period. The positive correlation result between the first vibration ratio and the first vibration exponent value of the target acquisition period is recorded as the vibration wear confidence level of the target acquisition period.

[0072] The number of negative values ​​among the first slope of grinding sound, the first slope of current, and the first slope of vibration in the target acquisition cycle is recorded as the number of descents in the target acquisition cycle.

[0073] The average of the confidence scores for noise wear, current wear, and vibration wear during the target acquisition cycle is denoted as the wear confidence score for the target acquisition cycle.

[0074] The greater the number of decreases in the target acquisition cycle and the higher the wear confidence level, the greater the likelihood that the steel ball has experienced wear during the target acquisition cycle and that the wear is significant.

[0075] When steel balls experience severe wear, their volume and mass decrease significantly. The kinetic energy of collisions between steel balls and between the steel balls and the cylinder decreases, and the impact becomes less sharp. Consequently, the number of high-frequency transient fluctuations in the acquired vibration signal decreases. Kurtosis is typically used to reflect the impact characteristics of the vibration signal; therefore, changes in kurtosis can be used to evaluate the degree of steel ball wear.

[0076] The vibration sequence of the target acquisition period is divided into equal parts. For each vibration subsequence, the kurtosis of the subsequence is calculated. A linear fit is then performed on the kurtosis of all subsequences divided from the vibration sequence of the target acquisition period. The normalized slope of the fitted line is recorded as the kurtosis slope of the target acquisition period. The range of kurtosis among all subsequences divided from the vibration sequence of the target acquisition period is recorded as the kurtosis range of the target acquisition period. The ratio of the kurtosis range to the kurtosis slope of the target acquisition period is recorded as the kurtosis ratio of the target acquisition period.

[0077] The calculation of kurtosis of the sequence is a well-known technique and will not be elaborated further. The first preset quantity is represented by 10 in this embodiment. In this embodiment, the sigmoid function is used to calculate the normalized value of the slope of the fitted line. The sigmoid function is a well-known technology and will not be described in detail here. As other implementation methods, implementers can use other methods of the prior art, such as the tanh function.

[0078] The larger the kurtosis range, the more significant the change in impact force caused by the steel ball collision within the acquisition period corresponding to the kurtosis range, and the more severe the wear on the steel ball. The kurtosis slope represents the trend of kurtosis variation among all vibration subsequences divided from the vibration sequence within the acquisition period corresponding to the kurtosis slope.

[0079] The positive correlation between the number of drops, wear confidence, and kurtosis ratio in the target acquisition cycle is recorded as the steel ball wear coefficient of the target acquisition cycle.

[0080] It is understandable that positive correlation processing is applied to the number of descents, wear confidence, and kurtosis ratio of the target acquisition cycle, that is, to ensure that the number of descents, wear confidence, and kurtosis ratio of the target acquisition cycle are positively correlated with the steel ball wear coefficient of the target acquisition cycle.

[0081] Preferably, as an embodiment of this application, the product of the number of drops in the target acquisition cycle, the wear confidence level, and the kurtosis ratio is denoted as the steel ball wear coefficient of the target acquisition cycle.

[0082] The steel ball wear coefficient is used to evaluate the degree of wear of the steel ball within the corresponding data collection period. The larger the steel ball wear coefficient, the greater the degree of wear of the steel ball within the corresponding data collection period. In this case, the wear of the steel ball is less matched with the preset rotation speed, and the greater the impact of steel ball wear on the grinding efficiency of the ball mill. Therefore, the corresponding rotation speed should be increased to improve the grinding efficiency of the ball mill.

[0083] The same method can be used to obtain the steel ball wear coefficient for any collection cycle.

[0084] At this point, the wear coefficient of the steel ball for all collection cycles has been obtained.

[0085] Step S003: Determine the wear difference coefficient of the target acquisition cycle based on the differences in current sequence, grinding sound sequence, and vibration sequence between the target acquisition cycle and the first acquisition cycle, as well as the difference in the steel ball wear coefficient between the target acquisition cycle and the first acquisition cycle.

[0086] Furthermore, by taking into account the difference in the degree of wear of the steel balls in each collection cycle compared to the first collection cycle, the error in evaluating the wear condition of the steel balls can be reduced.

[0087] The difference between the steel ball wear coefficient of the target acquisition cycle and that of the first acquisition cycle is denoted as the steel ball wear difference of the target acquisition cycle.

[0088] Steel ball wear variation is used to evaluate the difference in the degree of wear of steel balls in the target acquisition cycle relative to the first acquisition cycle. The greater the steel ball wear variation, the greater the difference in the degree of wear of steel balls in the target acquisition cycle relative to the first acquisition cycle. This also means a greater difference in the material-to-ball ratio inside the ball mill cylinder during the target acquisition cycle compared to the initial material-to-ball ratio at the time of grinding. Consequently, it becomes more necessary to adjust the rotational speed to adapt to the changed material-to-ball ratio in the ball mill and improve grinding efficiency.

[0089] The grinding sound sequence of the target acquisition period is processed using a sequence decomposition algorithm to obtain the trend sequence of the grinding sound sequence of the target acquisition period.

[0090] The trend sequence of the grinding sound sequence for each acquisition cycle can be obtained using the same method.

[0091] The absolute value of the similarity between the trend sequence of the target acquisition period and the grinding tone sequence of the first acquisition period is denoted as the initial trend similarity of the grinding tone in the target acquisition period.

[0092] The greater the initial trend similarity of the target acquisition cycle, the smaller the stability difference and the more significant the trend consistency between the target acquisition cycle and the first acquisition cycle of the grinding sound data.

[0093] The vibration sequence of the target acquisition period is processed using a sequence decomposition algorithm to obtain the trend sequence of the vibration sequence of the target acquisition period. The same method can be used to obtain the trend sequence of the vibration sequence of each acquisition period. The absolute value of the similarity between the trend sequences of the target acquisition period and the first acquisition period is denoted as the initial trend similarity of the vibration of the target acquisition period.

[0094] The current sequence of the target acquisition period is processed using a sequence decomposition algorithm to obtain the trend sequence of the current sequence of the target acquisition period. The same method can be used to obtain the trend sequence of the current sequence for each acquisition period. The absolute value of the similarity between the trend sequences of the current sequences of the target acquisition period and the first acquisition period is denoted as the initial trend similarity of the current in the target acquisition period.

[0095] The average of the initial trend similarity of the grinding sound, the initial trend similarity of the vibration, and the initial trend similarity of the current in the target acquisition cycle is denoted as the initial trend similarity of the target acquisition cycle.

[0096] In this embodiment, the STL time series decomposition algorithm is used as the sequence decomposition algorithm to obtain the trend sequence of the sequence, and the Pearson correlation coefficient is used to obtain the similarity between two trend sequences. As other implementation methods, in practical applications, in addition to achieving the purpose of sequence decomposition, implementers can use other methods in the prior art, such as the X11 seasonal decomposition algorithm, to perform sequence decomposition. In addition to achieving the purpose of sequence similarity evaluation, implementers can use other methods in the prior art, such as cosine similarity, to obtain the sequence similarity evaluation. This application does not impose any special limitations.

[0097] The normalized value of the ratio of the steel ball wear difference in the target acquisition cycle to the initial trend similarity is denoted as the wear difference coefficient of the target acquisition cycle.

[0098] In order to avoid the denominator being zero during the ratio calculation process, a preset value needs to be added to the denominator. In this example, the preset value is 0.005.

[0099] The wear difference coefficient is used to evaluate the degree of difference between the motion state of the steel balls in the ball mill during the corresponding sampling period and the motion state during the initial grinding. The larger the wear difference coefficient, the more severe the overall wear of the steel balls during the corresponding sampling period, the more significant the change in the material-to-ball ratio, and the greater the adjustment of the rotation speed.

[0100] The wear difference coefficient for any collection cycle can be obtained using the same method.

[0101] At this point, the wear difference coefficients for all acquisition cycles have been obtained.

[0102] Step S004: Based on the wear difference coefficient and rotational speed of the target acquisition cycle, the rotational speed of the ball mill for the composite mineral admixture is controlled.

[0103] Based on the wear difference coefficient and rotation rate of the target acquisition cycle, determine the adjustment rotation rate of the next adjacent acquisition cycle of the target acquisition cycle.

[0104] The product of the wear difference coefficient of the target acquisition cycle and the preset speed adjustment value is recorded as the adjustment amount of the target acquisition cycle. The sum of the adjustment amount of the target acquisition cycle and the speed is recorded as the adjustment speed of the next adjacent acquisition cycle of the target acquisition cycle.

[0105] It should be noted that in this embodiment, the rotation rate of the first acquisition cycle is directly assigned to 65%, and the rotation rate adjustment value is 7%. When the rotation rate of the next adjacent acquisition cycle of the target acquisition cycle is greater than 100%, the rotation rate of the next adjacent acquisition cycle of the target acquisition cycle is directly assigned to 100%.

[0106] By using the rotational speed of the target acquisition cycle as the value of the rotational speed of the next adjacent acquisition cycle, the rotational speed is matched with the gradually worn steel balls during the fine grinding process in the ball mill, thereby improving grinding efficiency, reducing the energy consumption of the ball mill's energy-saving motor, and improving the quality of the ground products.

[0107] Thus, ball mill control of composite mineral admixtures was achieved.

[0108] Based on the same inventive concept as the above method, this application embodiment also provides a control system for a composite mineral admixture ball mill, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described composite mineral admixture ball mill control methods.

[0109] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A control method for a ball mill containing composite mineral admixtures, characterized in that, The method includes the following steps: Drive current data, grinding sound data, and vibration data are collected at different times within different acquisition cycles to construct the current sequence, grinding sound sequence, and vibration sequence for each acquisition cycle. Any acquisition period is designated as the target acquisition period. Based on the changing trend of the grinding sound data within the grinding sound sequence of the target acquisition period, and the difference between the grinding sound sequence of the target acquisition period and the previous adjacent acquisition period, the first slope of the grinding sound and the grinding sound wear confidence level of the target acquisition period are determined. Following the method of determining the first slope of the grinding sound and the grinding sound wear confidence level of the target acquisition period based on the grinding sound sequence of the target acquisition period, the first slope of the current, the current wear confidence level, the first slope of the vibration, and the vibration wear confidence level of the target acquisition period are determined based on the current sequence and the vibration sequence of the target acquisition period, respectively. Combining the first slope of the grinding sound and the grinding sound wear confidence level, and the changing trend of the vibration data within the vibration sequence of the target acquisition period, the steel ball wear coefficient of the target acquisition period is determined. The wear difference coefficient for the target acquisition cycle is determined based on the differences in current sequence, grinding sound sequence, and vibration sequence between the target acquisition cycle and the first acquisition cycle, as well as the difference in steel ball wear coefficient between the target acquisition cycle and the first acquisition cycle. Based on the wear difference coefficient and rotation rate of the target collection cycle, the rotation rate of the ball mill for composite mineral admixtures is controlled. The method for determining the first slope of the grinding sound is as follows: A linear fit is performed on the grinding sound sequence of the target acquisition cycle, and the slope of the fitted line is denoted as the first slope of the grinding sound of the target acquisition cycle. The method for determining the confidence level of the grinding sound wear is as follows: The first slope of the grinding sound in the target acquisition cycle is taken as the exponent of the exponential function with the natural constant as the base, and the reciprocal of the calculated value of the exponential function is recorded as the first exponential value of the grinding sound in the target acquisition cycle. The first ratio of the grinding sound in the target acquisition period is determined based on the difference between the grinding sound sequence of the target acquisition period and the previous adjacent acquisition period. The positive correlation between the first ratio of grinding sound and the first index value of grinding sound in the target acquisition period is recorded as the grinding sound wear confidence level of the target acquisition period. The method for determining the steel ball wear coefficient for the target acquisition period by combining the first slope of the grinding sound and the confidence level of the grinding sound, as well as the changing trend of vibration data within the vibration sequence of the target acquisition period, includes the following specific methods: The number of negative values ​​among the first slope of grinding sound, the first slope of current, and the first slope of vibration in the target acquisition cycle is recorded as the number of descents in the target acquisition cycle. The average of the confidence scores for noise wear, current wear, and vibration wear during the target acquisition cycle is denoted as the wear confidence score for the target acquisition cycle. The vibration sequence of the target acquisition period is divided into a first preset number of vibration subsequences. The kurtosis of each vibration subsequence is calculated. Based on the changing trend of the kurtosis of all vibration subsequences divided from the vibration sequence of the target acquisition period, the kurtosis ratio of the target acquisition period is determined. The positive correlation between the number of drops, wear confidence, and kurtosis ratio in the target acquisition cycle is denoted as the steel ball wear coefficient in the target acquisition cycle. The method for determining the wear difference coefficient of the target acquisition cycle is as follows: The difference between the steel ball wear coefficient of the target acquisition cycle and that of the first acquisition cycle is denoted as the difference in steel ball wear coefficient of the target acquisition cycle. The initial trend similarity of the target acquisition cycle is determined based on the differences in current sequence, grinding sound sequence, and vibration sequence between the target acquisition cycle and the first acquisition cycle. The normalized value of the ratio of the difference in the steel ball wear coefficient during the target acquisition cycle to the initial trend similarity is denoted as the wear difference coefficient during the target acquisition cycle. The method for determining the initial trend similarity of the target acquisition period is as follows: Sequence decomposition is performed on the grinding sound sequence of the target acquisition period to obtain the trend sequence of the grinding sound sequence of the target acquisition period. The absolute value of the similarity between the trend sequence of the grinding sound sequence of the target acquisition period and the first acquisition period is denoted as the initial trend similarity of the grinding sound of the target acquisition period. The vibration sequence of the target acquisition period is decomposed to obtain the trend sequence of the vibration sequence of the target acquisition period; the absolute value of the similarity between the trend sequence of the vibration sequence of the target acquisition period and the first acquisition period is denoted as the initial trend similarity of the vibration of the target acquisition period. The current sequence of the target acquisition period is decomposed to obtain the trend sequence of the current sequence of the target acquisition period; the absolute value of the similarity between the trend sequence of the current sequence of the target acquisition period and the current sequence of the first acquisition period is denoted as the initial trend similarity of the current of the target acquisition period. The average of the initial trend similarity of the grinding sound, the initial trend similarity of the vibration, and the initial trend similarity of the current in the target acquisition cycle is denoted as the initial trend similarity of the target acquisition cycle. The method for determining the first ratio of the grinding sound in the target acquisition cycle is as follows: The ratio of the previous adjacent acquisition period of the target acquisition period to the mean value of the grinding sound data contained in the grinding sound sequence of the target acquisition period is denoted as the first grinding sound ratio of the target acquisition period. The method for determining the kurtosis ratio of the target acquisition period is as follows: The normalized value of the slope of the fitted line obtained by fitting the kurtosis of all vibration subsequences divided from the vibration sequence of the target acquisition period is denoted as the kurtosis slope of the target acquisition period. The range of kurtosis of all vibration subsequences divided from the vibration sequence of the target acquisition period is denoted as the kurtosis range of the target acquisition period. The ratio of the kurtosis range to the kurtosis slope of the target acquisition period is denoted as the kurtosis ratio of the target acquisition period.

2. The control method for a ball mill with composite mineral admixtures according to claim 1, characterized in that, The method for controlling the rotational speed of the ball mill for composite mineral admixtures based on the wear difference coefficient and rotational speed during the target collection cycle includes the following specific methods: The product of the wear difference coefficient of the target acquisition cycle and the preset speed adjustment value is recorded as the adjustment amount of the target acquisition cycle. The sum of the adjustment amount of the target acquisition cycle and the speed is recorded as the adjustment speed of the next adjacent acquisition cycle of the target acquisition cycle. The rotational speed of the target acquisition cycle is used as the rotational speed value of the next adjacent acquisition cycle.

3. A control system for a ball mill containing composite mineral admixtures, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as claimed in any one of claims 1-2.

Images