Mine equipment control system and method

Abstract

The invention discloses a mine equipment control system and method, and relates to the technical field of camping mine equipment control, and the method comprises the following steps: during the operation period of a crusher, obtaining various parameter information generated in the crushing process of the crusher in real time, including vibration response characteristics and energy transfer and dissipation characteristics; according to the method, the vibration response characteristics and the energy transfer characteristics of the crusher are obtained in real time, intelligent analysis is carried out in combination with a machine learning model, the system can efficiently recognize abnormal states and dynamically classify the abnormal states into accidental or non-accidental anomalies, and shutdown and false alarms are reduced. For accidental abnormity, the system automatically adjusts parameters to return to normal; and for non-accidental abnormity, the system triggers early warning and carries out maintenance, so that further damage is avoided. According to the method, the service life, the operation efficiency and the automation level of equipment are improved, and the maintenance cost and the sudden failure risk are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of mining equipment control, and in particular to a mining equipment control system and method. Background Art

[0002] Mining equipment control refers to the management, monitoring and adjustment of mechanical equipment and automation systems used in the mining production process to ensure the smooth progress of the production process. Mining equipment includes large machinery such as drills, crushers, conveyor belts, loaders, as well as sensors, monitoring systems and automation controllers used to control these equipment. Through precise control of equipment, mines can achieve efficient and stable production processes, reduce the occurrence of failures, optimize resource utilization, and ensure production safety. Equipment control also covers real-time monitoring of key parameters such as energy consumption, temperature, vibration, etc. The automation system or manual intervention can dynamically adjust the operating status of the equipment based on this data to avoid failures and production stagnation.

[0003] In mining production, the crusher plays a core role. It is responsible for crushing the collected large pieces of ore into small particles suitable for subsequent processing. The efficient operation of the crusher directly affects the continuity and stability of the entire production chain. Through the automatic crushing control system, the mine can automatically adjust the speed, opening size and power output of the crusher according to the hardness, size and other parameters of the ore to ensure the optimal efficiency of the crushing process. At the same time, the intelligent equipment control system relies on sensor technology, data acquisition systems and industrial automation software to help operators remotely monitor the operating status of the equipment and promptly detect and handle abnormal situations. Ultimately, this systematic equipment control helps to improve the production efficiency of the mine, reduce the risk of equipment damage, reduce energy consumption, and ensure the safety of operations.

[0004] The prior art has the following deficiencies:

[0005] The automatic crushing control system reflects the hardness of the ore through the vibration mode during the crushing process. However, the mining site is usually accompanied by the operation of other large equipment, which will generate a lot of background noise and ground vibration, which may affect the accuracy of the crusher vibration sensor. When the vibration sensor is disturbed, the system may have false alarms or data loss, resulting in failure of automatic control. Since the hardness and size of the ore are the core feedback parameters of the crushing control system, once these data are inaccurate, the system will not be able to dynamically adjust the speed, feed rate or crushing chamber opening size of the crusher, causing the crushing process to get out of control. In this case, the system may be forced to switch to manual mode, which directly affects the efficiency of mine automation, increases the frequency and difficulty of manual intervention, and reduces the stability and efficiency of overall production.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The purpose of the present invention is to provide a mining equipment control system and method. By acquiring the vibration response characteristics and energy transfer characteristics of the crusher in real time and combining them with machine learning models for intelligent analysis, the system can efficiently identify abnormal conditions and dynamically classify them into accidental or non-accidental abnormalities, thereby reducing downtime and false alarms. For accidental abnormalities, the system automatically adjusts the parameters to return to normal; for non-accidental abnormalities, the system triggers an early warning and performs maintenance to avoid further damage. This method improves the equipment life, operating efficiency and automation level, reduces maintenance costs and the risk of sudden failures, and solves the problems in the above-mentioned background technology.

[0008] In order to achieve the above object, the present invention provides the following technical solution: a mining equipment control method, comprising the following steps:

[0009] During the operation of the crusher, various parameter information generated during the crushing process of the crusher is obtained in real time, including vibration response characteristics and energy transfer and dissipation characteristics;

[0010] The obtained vibration response characteristics and energy transfer and dissipation characteristics are analyzed for abnormalities, and the vibration response characteristics and energy transfer and dissipation characteristics after abnormal analysis are input into the pre-learned machine learning model, and the amplitude fluctuation index is generated through the machine learning model, and the crusher operation process is intelligently evaluated through the amplitude fluctuation index;

[0011] Based on the evaluation results of the machine learning model, the operation process of the automated operation system is divided into normal operation state and abnormal operation state, and the abnormal crushing process of the crusher is intelligently perceived through the divided operation states;

[0012] According to the abnormal operation status evaluated by the machine learning model, the abnormal operation status of the automated operation system is further divided into accidental abnormality or non-accidental abnormality;

[0013] Take different countermeasures for different abnormalities.

[0014] Preferably, the vibration response characteristics include low-amplitude vibration response, and the energy transfer and dissipation characteristics include impact energy transfer and dissipation. The low-amplitude vibration response refers to the slight vibration reaction of the ore when it is subjected to limited external force during the crushing process, and the impact energy transfer and dissipation refers to the ability of the ore to absorb, transfer and dissipate external impact energy during the crushing process.

[0015] Preferably, a micro-amplitude response anomaly index is generated after an abnormal analysis of the low-amplitude vibration response, and an impact transfer dissipation fluctuation index is generated after an abnormal analysis of the impact energy transfer dissipation. The generated micro-amplitude response anomaly index and impact transfer dissipation fluctuation index are input into a pre-learned machine learning model, and an amplitude fluctuation index is generated through the machine learning model. The crusher operation process is intelligently evaluated through the amplitude fluctuation index.

[0016] Preferably, the logic of performing abnormal analysis on the low amplitude vibration response and generating the micro amplitude response abnormality index is as follows:

[0017] In the detection window, the vibration signal obtained from the vibration sensor in real time is band-pass filtered to eliminate high-frequency noise and low-frequency interference, retain the actual effective vibration frequency range in the crushing process, and calculate the filtered vibration signal. The calculation expression is as follows: Where V f (t) is the filtered vibration signal, which represents the filtered vibration signal at time t, V(t) is the vibration signal, which represents the vibration signal at time t, and f low is the lower frequency of the filter, f high is the upper frequency limit of the filter, is the bandpass filter function;

[0018] The filtered vibration signal is decomposed using the Hilbert-Huang transform to obtain a set of intrinsic modal components, which are calculated as follows: In the formula, C i (t) is the intrinsic modal component, which represents the i-th group of intrinsic modal components at time t, and N is the total number of intrinsic modal components;

[0019] For each natural modal component, the instantaneous frequency and instantaneous amplitude of the vibration signal are extracted using Hilbert transform. The calculation expression is as follows:

[0020]

[0021] , where A i (t) is the instantaneous amplitude of the vibration signal, which represents the instantaneous amplitude of the vibration signal of the i-th group of natural modal components at time t. is the Hilbert transform of the intrinsic mode components,

[0022]

[0023] , where ω i (t) is the instantaneous frequency of the vibration signal, which represents the instantaneous frequency of the vibration signal of the i-th group of natural modal components at time t, j is an imaginary unit, Represents the phase angle of the complex signal, the intrinsic modal component Ci (t) The phase at time t. The phase angle arg represents the phase information of the signal and is used to calculate the instantaneous frequency of the signal.

[0024] Based on the instantaneous frequency and amplitude of the vibration signal, the Markov model is used to capture the dynamic changes of the vibration signal. The state transition probability in the Markov model is defined and the calculation expression is as follows:

[0025]

[0026] , where P(S1→S2) is the state transition probability, which indicates the transition probability of the signal state from the normal state S1 to the abnormal state S2, and ∈ is the instantaneous amplitude A of the vibration signal i (t) and the instantaneous frequency ω of the vibration signal i (t) is the reference threshold, T is the length of the detection window;

[0027] Based on the instantaneous frequency ω of the vibration signal i (t) and the instantaneous amplitude A of the vibration signal i (t) and the state transition probability P(S1→S2) generate the micro-amplitude response anomaly index, and the calculation expression is as follows:

[0028]

[0029] , where MVR(t) represents the abnormal index of micro-amplitude response, k represents the total number of natural modal components, and λ is the regularization parameter. represents the time integral of the state transition probability from time 0 to time t.

[0030] Preferably, the logic of performing abnormal analysis on the impact energy transfer dissipation and generating the impact transfer dissipation fluctuation index is as follows:

[0031] In the detection window, the sensor collects the impact energy transfer and dissipation data of the ore during the crushing process in real time. These data include the input energy E in (t) and output energy E out (t), the energy transfer at each moment is recorded by discretizing the time series data, and the calculation expression is as follows: E(t) = E in (t)-E out (t), where E(t) is the net energy, which represents the energy not dissipated or transferred during the crushing process, and E in (t) is the input energy, which represents the total energy applied by the crusher to the ore at time t, E out (t) is the energy dissipated or converted in various ways during the crushing process of the ore;

[0032] The transient energy conversion characteristics in the preprocessed energy data are extracted in the form of nonlinear transformation. The transient energy at each time point t is calculated by combining Fourier transform with wavelet transform. The calculation expression is as follows:

[0033]

[0034] , where E transient (t) is the transient energy, which represents the instantaneous energy transfer characteristics at time t, ψ is the wavelet function, It is the scaling and translation of the wavelet function, τ is a time point in the past, s is the scale factor, and -∞ and ∞ represent the integration of time t from negative infinity to positive infinity;

[0035] Next, a nonlinear dynamic system model is constructed to characterize the transmission and dissipation of impact energy. The calculation expression is as follows:

[0036]

[0037] , where is the rate of energy change, α is the linear energy transfer coefficient, β is the nonlinear dissipation coefficient, γ is the coefficient related to the second-order time derivative, the weight of the acceleration term, αE(t) is the linear energy transfer term, βE 3 (t) is the nonlinear energy dissipation term, is the second-order time derivative term;

[0038] In order to analyze the abnormal behavior during the impact energy transfer process, it is necessary to detect the asymmetry in the energy transfer. By calculating the skewness coefficient of the dissipated energy, it is measured whether the energy transfer process has abnormal fluctuations or imbalances. The calculation expression is as follows:

[0039]

[0040] , where Skew(E) is the skewness coefficient of dissipated energy, μ is the equilibrium energy state of the system, that is, the long-term equilibrium point of energy transfer;

[0041] Comprehensive transient energy E transient (t), energy change rate And the skewness coefficient of dissipated energy Skew (E) to generate the shock transfer dissipation fluctuation index. The calculation expression is as follows:

[0042]

[0043] , where ΔE is the shock transmission dissipation fluctuation index.

[0044] Preferably, the micro-amplitude response abnormality index MVR(t) and the impact transmission dissipation fluctuation index ΔE are input into a pre-learned machine learning model, and the amplitude fluctuation index AF is generated by the machine learning model according to the formula:

[0045]

[0046] , where k1 and k2 are the preset proportional coefficients of the micro-amplitude response anomaly index MVR(t) and the impact transmission dissipation fluctuation index ΔE, and both k1 and k2 are greater than 0.

[0047] Preferably, the generated amplitude fluctuation index is compared and analyzed with a preset amplitude fluctuation index reference threshold, and the analysis result is as follows:

[0048] If the amplitude fluctuation index is greater than or equal to a preset amplitude fluctuation index reference threshold, the operation process of the automation operation system is classified as an abnormal operation state;

[0049] If the amplitude fluctuation index is less than a preset amplitude fluctuation index reference threshold, the operation process of the automation operation system is classified as a normal operation state.

[0050] Preferably, when an abnormal operation state is generated during the operation of the automatic operation system, a comprehensive analysis is performed on several amplitude fluctuation indexes subsequently generated by the operation of the automatic operation system and a preset reference threshold value of the amplitude fluctuation index to generate an operation evaluation coefficient YX XS , based on the formula: Where AF r represents the amplitude fluctuation index greater than or equal to the amplitude fluctuation index reference threshold, r represents the number of the amplitude fluctuation index greater than or equal to the amplitude fluctuation index reference threshold, r=1, 2, 3, ..., e, e is a positive integer;

[0051] The generated evaluation coefficient is compared and analyzed with the preset evaluation coefficient reference threshold, and the analysis results are as follows:

[0052] If the evaluation coefficient is greater than or equal to a preset evaluation coefficient reference threshold, the abnormal operation state of the automated operation system is further classified as a non-accidental abnormality;

[0053] If the evaluation coefficient is less than a preset evaluation coefficient reference threshold, the abnormal operation state of the automation operation system is further classified as an accidental abnormality.

[0054] A mining equipment control system includes a parameter real-time monitoring module, an abnormality analysis and fluctuation assessment module, an operation status identification module, an abnormality classification module, and an intelligent response and processing module:

[0055] The real-time parameter monitoring module obtains the various parameter information generated during the crushing process of the crusher in real time during the operation of the crusher, including vibration response characteristics and energy transfer and dissipation characteristics;

[0056] The abnormal analysis and fluctuation assessment module performs abnormal analysis on the acquired vibration response characteristics and energy transfer and dissipation characteristics, and inputs the vibration response characteristics and energy transfer and dissipation characteristics after abnormal analysis into the pre-learned machine learning model, generates the amplitude fluctuation index through the machine learning model, and performs intelligent evaluation on the crusher operation process through the amplitude fluctuation index;

[0057] The operation status recognition module divides the operation process of the automated operation system into normal operation state and abnormal operation state based on the evaluation results of the machine learning model, and intelligently perceives the abnormal crushing process of the crusher through the divided operation states;

[0058] The abnormality classification module further classifies the abnormal operation status of the automation system into accidental abnormality or non-accidental abnormality based on the abnormal operation status evaluated by the machine learning model;

[0059] The intelligent response and processing module takes different measures for different anomalies.

[0060] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0061] The present invention obtains the vibration response characteristics and energy transfer and dissipation characteristics of the crusher in real time during operation, and performs intelligent analysis based on the machine learning model. The system can efficiently identify abnormal conditions in the operation of the crusher, especially effectively detect faults when there is external interference or sensor inaccuracy. The amplitude fluctuation index combines the intelligent evaluation of the micro-amplitude response abnormality index and the impact transfer dissipation fluctuation index, so that the system can dynamically determine whether the operating status is abnormal, and further subdivide the abnormality into accidental abnormalities or non-accidental abnormalities. This classification helps to reduce unnecessary downtime and false alarms. Especially for accidental abnormalities, the system can automatically adjust parameters according to the fluctuation index to quickly restore the equipment to normal operation and avoid serious interference with the production process. This method reduces the frequency of manual monitoring and intervention, improves the automation level of the system, and ensures the continuity and efficiency of mine production.

[0062] For non-accidental anomalies, the system of the present invention takes strict countermeasures, including automatically triggering early warnings, suspending equipment operation, and conducting detailed diagnostic and maintenance procedures. This mechanism can effectively prevent the equipment from continuing to operate in a faulty state, reducing the risk of further damage to the equipment. In addition, combined with historical data analysis and diagnosis, the system can foresee and handle possible hardware failures and system problems in advance to ensure the stability of subsequent operations. This can not only extend the service life of the equipment and reduce long-term maintenance costs, but also improve the reliability and safety of the entire mine automation control system, and reduce the risk of sudden downtime and large-scale production losses. Through this preventive maintenance mechanism, the system can provide higher protection for mine operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0064] Figure 1 The present invention is a method flow chart of a mining equipment control method.

[0065] Figure 2 The figure is a module schematic diagram of a mining equipment control system of the present invention. DETAILED DESCRIPTION

[0066] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0067] The present invention provides Figure 1 A mining equipment control method shown includes the following steps:

[0068] During the operation of the crusher, various parameter information generated during the crushing process of the crusher is obtained in real time, including vibration response characteristics and energy transfer and dissipation characteristics;

[0069] The vibration response characteristics include low-amplitude vibration response, and the energy transfer and dissipation characteristics include impact energy transfer and dissipation. Low-amplitude vibration response refers to the slight vibration response of the ore when it is subjected to limited external forces during the crushing process. Impact energy transfer and dissipation refers to the ability of the ore to absorb, transfer and dissipate external impact energy during the crushing process.

[0070] The obtained vibration response characteristics and energy transfer and dissipation characteristics are analyzed for abnormalities, and the vibration response characteristics and energy transfer and dissipation characteristics after abnormal analysis are input into the pre-learned machine learning model, and the amplitude fluctuation index is generated through the machine learning model, and the crusher operation process is intelligently evaluated through the amplitude fluctuation index;

[0071] After abnormal analysis of low-amplitude vibration response, a micro-amplitude response abnormality index is generated. After abnormal analysis of impact energy transfer dissipation, an impact transfer dissipation fluctuation index is generated. The generated micro-amplitude response abnormality index and impact transfer dissipation fluctuation index are input into a pre-learned machine learning model. The amplitude fluctuation index is generated through the machine learning model. The crusher operation process is intelligently evaluated through the amplitude fluctuation index.

[0072] Changes in the low-amplitude vibration response during the crushing process of the crusher may indicate that the vibration sensor is disturbed, resulting in false alarms or data loss in the system, which in turn makes the automatic control ineffective. The low-amplitude vibration response usually reflects the vibration characteristics of the ore in the early stage of crushing or under slight stress. When it changes abnormally, it may mean that the sensor is receiving signal interference generated by external background noise, ground vibration or mechanical equipment self-vibration, which is not generated during the ore crushing process. These interferences will affect the sensor's accurate collection of the actual vibration data of the ore, causing the system to misjudge the hardness and crushing state of the ore. If the system automatically adjusts based on incorrect vibration feedback, it may incorrectly adjust parameters such as the speed, feed rate or impact force of the crusher, thereby disrupting the normal crushing process. Under long-term interference, the system cannot effectively distinguish between the true characteristics of the ore and noise interference, and the automatic control may frequently make mistakes, forcing the system to switch to manual operation mode, reducing production efficiency and increasing the maintenance and manual intervention requirements of the equipment. Therefore, abnormal changes in low-amplitude vibration response are often important signals of sensor interference, which directly affect the automation performance and stability of the crushing control system.

[0073] The logic of abnormal analysis of low amplitude vibration response and generation of micro amplitude response abnormal index is as follows:

[0074] In the detection window, the vibration signal obtained from the vibration sensor in real time is band-pass filtered to eliminate high-frequency noise and low-frequency interference, retain the actual effective vibration frequency range in the crushing process, and calculate the filtered vibration signal. The calculation expression is as follows: Where V f (t) is the filtered vibration signal, which represents the filtered vibration signal at time t, V(t) is the vibration signal, which represents the vibration signal at time t, and f low is the lower frequency of the filter, f high is the upper frequency limit of the filter, is the bandpass filter function;

[0075] It should be noted that band-pass filtering is a signal processing technique that allows signals within a specific frequency range to pass through, while blocking signals below or above that frequency range. In other words, the band-pass filter filters the input signal, retaining only the signal within the preset frequency range, and removing low-frequency and high-frequency noise and interference.

[0076] The role of bandpass filtering:

[0077] 1. Remove noise and interference: The main function of the bandpass filter is to eliminate unnecessary frequency components in the signal, especially low-frequency noise (such as DC offset or background vibration) and high-frequency noise (such as equipment electromagnetic interference). This is very important for maintaining the clarity and accuracy of the signal.

[0078] 2. Extract useful signals: In the vibration analysis of the crusher, the vibration signal usually contains many frequency components, and only certain frequency ranges are related to ore crushing. Through bandpass filtering, the system can extract effective frequency information related to equipment operation or ore hardness from the complex signal and ignore irrelevant frequency components.

[0079] 3. Signal conditioning and enhancement: Bandpass filters can enhance signals in certain frequency bands and weaken other frequency bands, thereby improving the accuracy and effectiveness of analysis. For example, in vibration monitoring of mechanical equipment, a specific frequency range reflects the normal operation of the equipment, while abnormal frequencies may indicate equipment failure.

[0080] In summary, the role of bandpass filtering is to improve the effectiveness of signal analysis by isolating the signal within the target frequency range, helping to eliminate interference and enhancing the ability to identify specific events or phenomena.

[0081] The filtered vibration signal is decomposed using the Hilbert-Huang transform (HHT) to obtain a set of intrinsic modal components, which are calculated as follows: In the formula, C i (t) is the intrinsic modal component, which represents the i-th group of intrinsic modal components at time t, and N is the total number of intrinsic modal components;

[0082] It should be noted that the intrinsic mode function (IMF) is a number of simple vibration components decomposed from a complex signal, and each IMF represents the vibration characteristics of a specific frequency range in the signal. These components are obtained through the Hilbert-Huang Transform (HHT), which can accurately capture the local frequency and amplitude changes in the signal. Each intrinsic mode component is an independent mode of the signal, representing the characteristics of the signal at different frequencies.

[0083] The reason for using the Hilbert-Huang transform to decompose the filtered vibration signal is that traditional frequency analysis methods (such as Fourier transform) are only applicable to stationary signals, while most actual vibration signals, such as the vibration during crusher operation, are often nonlinear and non-stationary. HHT can process these complex signals and decompose them into multiple intrinsic modal components to capture the different frequency characteristics of the signal over time. This decomposition method can provide more refined time-frequency analysis, help identify abnormal features in the vibration signal, and thus better understand the dynamic behavior of the crushing process.

[0084] For each natural modal component, the instantaneous frequency and instantaneous amplitude of the vibration signal are extracted using Hilbert transform. The calculation expression is as follows:

[0085]

[0086] , where A i (t) is the instantaneous amplitude of the vibration signal, which represents the instantaneous amplitude of the vibration signal of the i-th group of natural modal components at time t. is the Hilbert transform of the intrinsic mode components,

[0087]

[0088] , where ω i (t) is the instantaneous frequency of the vibration signal, which represents the instantaneous frequency of the vibration signal of the i-th group of natural modal components at time t, j is an imaginary unit, Represents the phase angle of the complex signal, the intrinsic modal component C i (t) The phase at time t. The phase angle arg represents the phase information of the signal and is used to calculate the instantaneous frequency of the signal.

[0089] It should be noted that the phase angle refers to the relationship between the phase of a signal and time in a complex number-represented signal. The role of the phase angle is that it can describe the time-domain variation characteristics of the signal, especially the instantaneous phase and instantaneous frequency of the signal. By analyzing the rate of change of the phase angle over time (i.e., the derivative of the phase angle), the instantaneous frequency of the signal can be obtained, reflecting the local frequency variation of the signal. Instantaneous phase information is very important for processing nonlinear and non-stationary signals, because the frequency of these signals changes over time, and traditional frequency domain analysis methods (such as Fourier transform) cannot capture these dynamic changes. Phase angle is widely used in signal processing, communication, vibration analysis and other fields, especially for understanding the time-frequency characteristics of signals.

[0090] Based on the instantaneous frequency and amplitude of the vibration signal, the Markov model is used to capture the dynamic changes of the vibration signal. The state transition probability in the Markov model is defined and the calculation expression is as follows:

[0091]

[0092] , where P(S1→S2) is the state transition probability, which indicates the transition probability of the signal state from the normal state S1 to the abnormal state S2, and ∈ is the instantaneous amplitude A of the vibration signal i (t) and the instantaneous frequency ω of the vibration signal i (t) is the reference threshold, T is the length of the detection window;

[0093] It should be noted that the reason why the instantaneous amplitude and instantaneous frequency share the same threshold ∈ is that they jointly reflect the local characteristics of the signal at a specific time point, and in anomaly detection, the anomalies of amplitude and frequency are often related. Specifically, when the amplitude of the signal vibration becomes abnormally large, the frequency may also change significantly, and vice versa. Therefore, by sharing a threshold ∈, the system can capture significant changes in vibration amplitude and frequency at the same time to identify abnormal signals.

[0094] Detailed reasons:

[0095] 1. Linkage relationship: In physical phenomena, the instantaneous amplitude and instantaneous frequency of a signal are usually interrelated. When a vibration system fails, is disturbed by external factors, or when the signal characteristics suddenly change, changes in amplitude and frequency often occur simultaneously. For example, in mechanical fault detection, a large vibration amplitude may be accompanied by frequency fluctuations, reflecting the unstable operation of the equipment. Setting a single threshold can simplify the detection logic and avoid the complexity of setting independent thresholds for amplitude and frequency.

[0096] 2. Comprehensive effect of threshold: A reasonably set threshold ∈ can play a role in comprehensive evaluation and detect any abnormalities in amplitude and frequency. If either the instantaneous amplitude or the instantaneous frequency exceeds this threshold, it indicates that the overall dynamic behavior of the signal has changed significantly, whether it is an abnormally large amplitude or a sudden change in frequency, the system can make corresponding adjustments or alarms through this threshold.

[0097] Therefore, the design of a common threshold ∈ can simplify the implementation of the algorithm, while ensuring that the system can keenly capture any significant changes in vibration characteristics in the signal during the monitoring process, effectively improving the efficiency of anomaly detection.

[0098] Based on the instantaneous frequency ω of the vibration signal i (t) and the instantaneous amplitude A of the vibration signal i (t) and the state transition probability P(S1→S2) generate the micro-amplitude response anomaly index, and the calculation expression is as follows:

[0099]

[0100] , where MVR(t) represents the abnormal index of micro-amplitude response, k represents the total number of natural modal components, and λ is the regularization parameter. represents the time integral of the state transition probability from time 0 to time t;

[0101] It should be noted that the regularization parameter is an adjustment coefficient in machine learning and statistical models, which is used to control the complexity of the model and prevent the model from overfitting or instability during processing. In optimization problems, the regularization parameter constrains the weights or coefficients of the model by adding a penalty term to the loss function, so that the model will not overfit the details when processing training data, but will be more generalized. In physical systems and signal processing, the regularization parameter can balance the weights of multiple influencing factors. For example, in vibration analysis, it is used to balance the impact between instantaneous fluctuations and long-term trends, ensuring that the model is sensitive enough to short-term anomalies while not ignoring long-term cumulative changes.

[0102] From the calculation expression of the micro-amplitude response anomaly index, it can be seen that the larger the micro-amplitude response anomaly index performance value generated after the abnormal analysis of the low-amplitude vibration response, the higher the possibility of abnormal fluctuations in the low-amplitude vibration response, indicating that the vibration sensor may be subject to external interference, such as noise or mechanical vibration interference, which will increase the probability of system false alarms or data loss, and may eventually lead to the failure of automatic control. On the contrary, the smaller the performance value of the micro-amplitude response anomaly index, the more stable the vibration signal, the less interference, the lower the risk of system false alarms and data loss, and the normal operation of automatic control.

[0103] During the crushing process of the crusher, the change of impact energy transfer dissipation does not match the actual ore characteristics or cannot be accurately captured by the system, which may cause system false alarms or data loss, thereby affecting the normal operation of the automatic control. The impact energy transfer dissipation characteristics reflect the ability of the ore to absorb and transfer energy during crushing, which is closely related to the hardness, density and deformation characteristics of the ore. If the vibration sensor cannot accurately capture the real vibration data during the crushing process due to external interference (such as mechanical noise, ground vibration or mechanical failure of the equipment itself), the system may misjudge the hardness or impact energy transfer characteristics of the ore. This misjudgment may cause the crusher to be unable to adjust parameters such as impact intensity, rotation speed or feed rate in time, and the crushing process will be out of control. The system may switch to manual mode or shut down. Especially in the automatic crushing control system that is highly dependent on real-time feedback, vibration data is one of the core parameters. Inaccurate data will directly affect the control strategy of the system and cause the failure of automatic control.

[0104] The logic of abnormal analysis of impact energy transfer dissipation and generating the impact transfer dissipation fluctuation index is as follows:

[0105] In the detection window, the sensor collects the impact energy transfer and dissipation data of the ore during the crushing process in real time. These data include the input energy E in (t) and output energy E out (t), the energy transfer at each moment is recorded by discretizing the time series data, and the calculation expression is as follows: E(t) = E in (t)-E out (t), where E(t) is the net energy, which represents the energy not dissipated or transferred during the crushing process, and E in (t) is the input energy, which represents the total energy applied by the crusher to the ore at time t, E out (t) is the energy dissipated or converted in various ways during the crushing process of the ore;

[0106] The transient energy conversion characteristics in the preprocessed energy data are extracted in the form of nonlinear transformation. The transient energy at each time point t is calculated by combining Fourier transform with wavelet transform. The calculation expression is as follows:

[0107]

[0108] , where E transient (t) is the transient energy, which represents the instantaneous energy transfer characteristics at time t, ψ is the wavelet function, It is the scaling and translation of the wavelet function, τ is a time point in the past, s is the scale factor, and -∞ and ∞ represent the integration of time t from negative infinity to positive infinity;

[0109] The scale factor is a key parameter in the wavelet transform, which controls the stretching or compression degree of the wavelet function and is used to adjust the time scale of signal analysis. A larger scale factor will stretch the wavelet, which is suitable for capturing the low-frequency and long-term characteristics of the signal and helping to analyze slow changes or smooth trends; a smaller scale factor will compress the wavelet and focus on high-frequency and short-term characteristics, which is used to detect fast-changing and sharp signal fluctuations. By changing the scale factor, the wavelet transform can perform multi-resolution analysis of the signal at different time scales and capture the local and global characteristics of the signal.

[0110] The reason why Fourier transform combined with wavelet transform is used to calculate the transient energy at each time point is that these two transforms have their own unique advantages. Their combination can make up for the limitations of a single transform and provide a more comprehensive signal analysis, especially in nonlinear and non-stationary signals. The specific advantages are as follows:

[0111] 1. Advantages of Fourier Transform

[0112] Powerful frequency domain analysis: Fourier transform can convert time domain signals into frequency domain signals, thereby accurately analyzing the frequency components of the signal. For periodic signals or signals with stable frequencies, Fourier transform provides powerful spectrum information, making the frequency components clear at a glance.

[0113] Linearity and stability: Fourier transform works very well for linear, stationary signals and can accurately extract frequency characteristics.

[0114] 2. Advantages of wavelet transform

[0115] Time-frequency localization: Wavelet transform can provide time and frequency information at the same time, which makes it possible to perform local analysis on non-stationary signals. Wavelet transform can perform multi-resolution analysis on different time scales, that is, it can simultaneously obtain high-precision time and frequency information in different frequency ranges.

[0116] Nonlinear signal processing: Wavelet transform is very suitable for processing nonlinear and non-stationary signals, such as energy fluctuations in the fragmentation process. It can effectively capture mutations and transient phenomena in the signal.

[0117] 3. Advantages of combining Fourier transform with wavelet transform

[0118] Combination of global and local: Fourier transform is good at capturing the global frequency characteristics of the signal, while wavelet transform can provide detailed information on local time and frequency. The combination of the two can capture instantaneous signal changes, that is, transient energy, while maintaining frequency domain accuracy. Fourier transform is used to analyze the frequency characteristics of the entire signal, while wavelet transform further distinguishes short-term fluctuation characteristics based on time and frequency localization.

[0119] Processing complex signals: Fourier transform combined with wavelet transform can analyze both stationary and non-stationary signals. This is particularly important in the analysis of energy transfer in nonlinear processes such as crushers, because these processes often contain short-term violent fluctuations (transient changes) and long-term stable behavior.

[0120] Multi-resolution analysis: By adjusting the scale factor of the wavelet transform, the system can flexibly process signals at different time scales, while the Fourier transform provides a global frequency context. When used together, you can start with the global spectrum and then use the wavelet transform to focus locally on the energy changes at a specific time point.

[0121] The combination of Fourier transform and wavelet transform has a powerful analytical ability, and can capture global frequency characteristics and local transient characteristics at the same time. This combination not only makes up for the lack of local information in time of Fourier transform, but also makes up for the limitations of wavelet transform in global frequency analysis. It is very suitable for transient energy calculation of complex, nonlinear and non-stationary signals, such as the dynamic change of energy during the operation of the crusher.

[0122] Next, a nonlinear dynamic system model is constructed to characterize the transmission and dissipation of impact energy. The calculation expression is as follows:

[0123]

[0124] , where is the rate of energy change, α is the linear energy transfer coefficient, β is the nonlinear dissipation coefficient, γ is the coefficient related to the second-order time derivative, the weight of the acceleration term, αE(t) is the linear energy transfer term, βE 3 (t) is the nonlinear energy dissipation term, is the second-order time derivative term;

[0125] In order to analyze the abnormal behavior during the impact energy transfer process, it is necessary to detect the asymmetry in the energy transfer. By calculating the skewness coefficient of the dissipated energy, it is measured whether the energy transfer process has abnormal fluctuations or imbalances. The calculation expression is as follows:

[0126]

[0127] , where Skew(E) is the skewness coefficient of dissipated energy, μ is the equilibrium energy state of the system, that is, the long-term equilibrium point of energy transfer;

[0128] The skewness coefficient can capture abnormal mutations in the dissipation process and is a key indicator for further analyzing system anomalies;

[0129] Comprehensive transient energy E transient (t), energy change rate And the skewness coefficient of dissipated energy Skew (E) to generate the shock transfer dissipation fluctuation index. The calculation expression is as follows:

[0130]

[0131] , where ΔE is the shock transmission dissipation fluctuation index;

[0132] From the calculation expression of the shock transfer dissipation fluctuation index, it can be seen that the larger the shock transfer dissipation fluctuation index performance value generated after the abnormal analysis of the shock energy transfer dissipation, the greater the interference to the vibration sensor, the more violent the fluctuation in the energy transfer process, the greater the probability of false alarm or data loss in the system, and thus the risk of automatic control failure. On the contrary, if the shock transfer dissipation fluctuation index performance value is smaller, it means that the energy transfer process of the system is relatively stable, the vibration sensor is less disturbed, the risk of false alarm or data loss is reduced, and the probability of automatic control failure is also reduced.

[0133] The micro-amplitude response anomaly index MVR(t) and the impact transmission dissipation fluctuation index ΔE are input into the pre-learned machine learning model, and the amplitude fluctuation index AF is generated by the machine learning model according to the formula:

[0134]

[0135] , where k1 and k2 are the preset proportional coefficients of the micro-amplitude response anomaly index MVR(t) and the impact transmission dissipation fluctuation index ΔE, and both k1 and k2 are greater than 0.

[0136] It can be seen from the calculation expression of the amplitude fluctuation index that the larger the micro-amplitude response abnormality index performance value generated after the abnormal analysis of the low-amplitude vibration response, and the larger the impact transfer dissipation fluctuation index performance value generated after the abnormal analysis of the impact energy transfer dissipation, the larger the amplitude fluctuation index performance value, indicating that the vibration sensor may be subject to external interference, the system false alarm or data loss, and the greater the probability of failure of the automation system. Otherwise, it means that the vibration sensor is relatively less disturbed.

[0137] Based on the evaluation results of the machine learning model, the operation process of the automated operation system is divided into normal operation state and abnormal operation state, and the abnormal crushing process of the crusher is intelligently perceived through the divided operation states;

[0138] The generated amplitude fluctuation index is compared and analyzed with the preset amplitude fluctuation index reference threshold. The analysis results are as follows:

[0139] If the amplitude fluctuation index is greater than or equal to a preset amplitude fluctuation index reference threshold, the operation process of the automation operation system is classified as an abnormal operation state;

[0140] If the amplitude fluctuation index is less than a preset amplitude fluctuation index reference threshold, the operation process of the automation operation system is classified as a normal operation state;

[0141] According to the abnormal operation status evaluated by the machine learning model, the abnormal operation status of the automated operation system is further divided into accidental abnormality or non-accidental abnormality;

[0142] When an abnormal operation state is generated during the operation of the automatic operation system, a comprehensive analysis is performed on several amplitude fluctuation indexes generated by the subsequent operation of the automatic operation system and the preset amplitude fluctuation index reference threshold to generate the operation evaluation coefficient YX XS , based on the formula: Where AF r represents the amplitude fluctuation index greater than or equal to the amplitude fluctuation index reference threshold, r represents the number of the amplitude fluctuation index greater than or equal to the amplitude fluctuation index reference threshold, r=1, 2, 3, ..., e, e is a positive integer;

[0143] The generated evaluation coefficient is compared and analyzed with the preset evaluation coefficient reference threshold, and the analysis results are as follows:

[0144] If the evaluation coefficient is greater than or equal to a preset evaluation coefficient reference threshold, the abnormal operation state of the automated operation system is further classified as a non-accidental abnormality;

[0145] If the evaluation coefficient is less than a preset evaluation coefficient reference threshold, the abnormal operation state of the automation operation system is further classified as an accidental abnormality;

[0146] The system automatically operates by taking different countermeasures for different anomalies.

[0147] For occasional anomalies, it is usually not necessary to take drastic intervention measures. You can adjust the operating parameters, such as reducing the operating intensity of the equipment or temporarily suspending the operation, to restore the system to normal state. At the same time, monitor the amplitude fluctuation index generated subsequently to confirm whether the problem persists. If the anomaly does not recur, the system can gradually return to normal operating mode. Such anomalies are generally caused by short-term external interference and will not cause long-term impact;

[0148] For non-accidental anomalies, strict response measures should be taken. First, an early warning should be automatically triggered to suspend or significantly slow down the operation of related equipment to avoid further damage. Then, a detailed diagnosis and maintenance procedure should be arranged, including checking the operating status of sensors, equipment components, and the entire system to eliminate hardware failures or persistent problems. Finally, the system historical data should be analyzed to ensure future operational stability.

[0149] The present invention obtains the vibration response characteristics and energy transfer and dissipation characteristics of the crusher in real time during operation, and performs intelligent analysis based on the machine learning model. The system can efficiently identify abnormal conditions in the operation of the crusher, especially effectively detect faults when there is external interference or sensor inaccuracy. The amplitude fluctuation index combines the intelligent evaluation of the micro-amplitude response abnormality index and the impact transfer dissipation fluctuation index, so that the system can dynamically determine whether the operating status is abnormal, and further subdivide the abnormality into accidental abnormalities or non-accidental abnormalities. This classification helps to reduce unnecessary downtime and false alarms. Especially for accidental abnormalities, the system can automatically adjust parameters according to the fluctuation index to quickly restore the equipment to normal operation and avoid serious interference with the production process. This method reduces the frequency of manual monitoring and intervention, improves the automation level of the system, and ensures the continuity and efficiency of mine production.

[0150] For non-accidental anomalies, the system of the present invention takes strict countermeasures, including automatically triggering early warnings, suspending equipment operation, and conducting detailed diagnostic and maintenance procedures. This mechanism can effectively prevent the equipment from continuing to operate in a faulty state, reducing the risk of further damage to the equipment. In addition, combined with historical data analysis and diagnosis, the system can foresee and handle possible hardware failures and system problems in advance to ensure the stability of subsequent operations. This can not only extend the service life of the equipment and reduce long-term maintenance costs, but also improve the reliability and safety of the entire mine automation control system, and reduce the risk of sudden downtime and large-scale production losses. Through this preventive maintenance mechanism, the system can provide higher protection for mine operations.

[0151] The present invention provides Figure 2 A mining equipment control system shown in FIG. 1 includes a parameter real-time monitoring module, an abnormality analysis and fluctuation assessment module, an operation status identification module, an abnormality classification module, and an intelligent response and processing module:

[0152] The real-time parameter monitoring module obtains the various parameter information generated during the crushing process of the crusher in real time during the operation of the crusher, including vibration response characteristics and energy transfer and dissipation characteristics;

[0153] The abnormal analysis and fluctuation assessment module performs abnormal analysis on the acquired vibration response characteristics and energy transfer and dissipation characteristics, and inputs the vibration response characteristics and energy transfer and dissipation characteristics after abnormal analysis into the pre-learned machine learning model, generates the amplitude fluctuation index through the machine learning model, and performs intelligent evaluation on the crusher operation process through the amplitude fluctuation index;

[0154] The operation status recognition module divides the operation process of the automated operation system into normal operation state and abnormal operation state based on the evaluation results of the machine learning model, and intelligently perceives the abnormal crushing process of the crusher through the divided operation states;

[0155] The abnormality classification module further classifies the abnormal operation status of the automation system into accidental abnormality or non-accidental abnormality based on the abnormal operation status evaluated by the machine learning model;

[0156] The intelligent response and processing module takes different measures for different anomalies.

[0157] A mining equipment control method provided in an embodiment of the present invention is implemented by the above-mentioned mining equipment control system. The specific method and process of a mining equipment control system are detailed in the embodiment of the above-mentioned mining equipment control method, which will not be repeated here.

[0158] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A mining equipment control method, characterized in that: The following steps are involved: During the operation of the crusher, various parameter information generated during the crushing process of the crusher is obtained in real time, including vibration response characteristics and energy transfer and dissipation characteristics; The obtained vibration response characteristics and energy transfer and dissipation characteristics are analyzed for abnormalities, and the vibration response characteristics and energy transfer and dissipation characteristics after abnormal analysis are input into the pre-learned machine learning model, and the amplitude fluctuation index is generated through the machine learning model, and the crusher operation process is intelligently evaluated through the amplitude fluctuation index; Based on the evaluation results of the machine learning model, the operation process of the automated operation system is divided into normal operation state and abnormal operation state, and the abnormal crushing process of the crusher is intelligently perceived through the divided operation states; According to the abnormal operation status evaluated by the machine learning model, the abnormal operation status of the automated operation system is further divided into accidental abnormality or non-accidental abnormality; Take different countermeasures for different abnormalities.

2. A mining equipment control method according to claim 1, characterized in that: The vibration response characteristics include low-amplitude vibration response, and the energy transfer and dissipation characteristics include impact energy transfer and dissipation. Low-amplitude vibration response refers to the slight vibration response of the ore when it is subjected to limited external force during the crushing process. Impact energy transfer and dissipation refers to the ability of the ore to absorb, transfer and dissipate external impact energy during the crushing process.

3. A mining equipment control method according to claim 2, characterized in that: After performing an abnormal analysis on the low-amplitude vibration response, a micro-amplitude response abnormality index is generated. After performing an abnormal analysis on the impact energy transfer dissipation, an impact transfer dissipation fluctuation index is generated. The generated micro-amplitude response abnormality index and impact transfer dissipation fluctuation index are input into a pre-learned machine learning model, and an amplitude fluctuation index is generated through the machine learning model. The crusher operation process is intelligently evaluated through the amplitude fluctuation index.

4. A mining equipment control method according to claim 3, characterized in that: The logic of abnormal analysis of low amplitude vibration response and generation of micro amplitude response abnormal index is as follows: In the detection window, the vibration signal obtained from the vibration sensor in real time is band-pass filtered to eliminate high-frequency noise and low-frequency interference, retain the actual effective vibration frequency range in the crushing process, and calculate the filtered vibration signal. The calculation expression is as follows: Where V f (t) is the filtered vibration signal, which represents the filtered vibration signal at time t, V(t) is the vibration signal, which represents the vibration signal at time t, and f low is the lower frequency of the filter, f high is the upper frequency limit of the filter, is the bandpass filter function; The filtered vibration signal is decomposed using the Hilbert-Huang transform to obtain a set of intrinsic modal components, which are calculated as follows: In the formula, C i (t) is the intrinsic modal component, which represents the i-th group of intrinsic modal components at time t, and N is the total number of intrinsic modal components; For each natural modal component, the instantaneous frequency and instantaneous amplitude of the vibration signal are extracted using Hilbert transform. The calculation expression is as follows: ， In the formula, A i (t) is the instantaneous amplitude of the vibration signal, which represents the instantaneous amplitude of the vibration signal of the i-th group of natural modal components at time t. is the Hilbert transform of the intrinsic mode components, ， In the formula, ω i (t) is the instantaneous frequency of the vibration signal, which represents the instantaneous frequency of the vibration signal of the i-th group of natural modal components at time t, j is an imaginary unit, Represents the phase angle of the complex signal, the intrinsic modal component C i (t) The phase at time t. The phase angle arg represents the phase information of the signal and is used to calculate the instantaneous frequency of the signal. Based on the instantaneous frequency and amplitude of the vibration signal, the Markov model is used to capture the dynamic changes of the vibration signal. The state transition probability in the Markov model is defined and the calculation expression is as follows: ， Where P(S1→S2) is the state transition probability, which indicates the transition probability of the signal state from the normal state S1 to the abnormal state S2, and ∈ is the instantaneous amplitude A of the vibration signal. i (t) and the instantaneous frequency ω of the vibration signal i (t) is the reference threshold, T is the length of the detection window; Based on the instantaneous frequency ω of the vibration signal i (t) and the instantaneous amplitude A of the vibration signal i (t) and the state transition probability P(S1→S2) generate the micro-amplitude response anomaly index, and the calculation expression is as follows: ， Where MVR(t) represents the abnormal index of micro-amplitude response, k represents the total number of natural modal components, and λ is the regularization parameter. represents the time integral of the state transition probability from time 0 to time t.

5. A mining equipment control method according to claim 3, characterized in that: The logic of abnormal analysis of impact energy transfer dissipation and generating the impact transfer dissipation fluctuation index is as follows: In the detection window, the sensor collects the impact energy transfer and dissipation data of the ore in the crushing process in real time. These data include the input energy E in (t) and output energy E out (t), the energy transfer at each moment is recorded by discretizing the time series data, and the calculation expression is as follows: E(t) = E in (t)-E out (t), where E(t) is the net energy, which represents the energy not dissipated or transferred during the crushing process, and E in (t) is the input energy, which represents the total energy applied by the crusher to the ore at time t, E out (t) is the energy dissipated or converted in various ways during the crushing process of the ore; The transient energy conversion characteristics in the preprocessed energy data are extracted in the form of nonlinear transformation. The transient energy at each time point t is calculated by combining Fourier transform with wavelet transform. The calculation expression is as follows: ， In the formula, E transient (t) is the transient energy, which represents the instantaneous energy transfer characteristics at time t, ψ is the wavelet function, It is the scaling and translation of the wavelet function, τ is a time point in the past, s is the scale factor, and -∞ and ∞ represent the integration of time t from negative infinity to positive infinity; Next, a nonlinear dynamic system model is constructed to characterize the transmission and dissipation of impact energy. The calculation expression is as follows: ， In the formula, is the rate of energy change, α is the linear energy transfer coefficient, β is the nonlinear dissipation coefficient, γ is the coefficient related to the second-order time derivative, the weight of the acceleration term, αE(t) is the linear energy transfer term, βE 3 (t) is the nonlinear energy dissipation term, is the second-order time derivative term; In order to analyze the abnormal behavior during the impact energy transfer process, it is necessary to detect the asymmetry in the energy transfer. By calculating the skewness coefficient of the dissipated energy, it is measured whether the energy transfer process has abnormal fluctuations or imbalances. The calculation expression is as follows: ， Where Skew(E) is the skewness coefficient of dissipated energy, μ is the equilibrium energy state of the system, that is, the long-term equilibrium point of energy transfer; Comprehensive transient energy E transient (t), energy change rate And the skewness coefficient of dissipated energy Skew (E) to generate the shock transfer dissipation fluctuation index. The calculation expression is as follows: ， Where ΔE is the shock transmission dissipation fluctuation index.

6. A mining equipment control method according to claim 3, characterized in that: The micro-amplitude response anomaly index MVR(t) and the impact transmission dissipation fluctuation index ΔE are input into the pre-learned machine learning model, and the amplitude fluctuation index AF is generated by the machine learning model according to the formula: ， Wherein, k1 and k2 are the preset proportional coefficients of the micro-amplitude response abnormality index MVR(t) and the impact transmission dissipation fluctuation index ΔE, and both k1 and k2 are greater than 0.

7. A mining equipment control method according to claim 6, characterized in that: The generated amplitude fluctuation index is compared and analyzed with the preset amplitude fluctuation index reference threshold. The analysis results are as follows: If the amplitude fluctuation index is greater than or equal to a preset amplitude fluctuation index reference threshold, the operation process of the automation operation system is classified as an abnormal operation state; If the amplitude fluctuation index is less than a preset amplitude fluctuation index reference threshold, the operation process of the automation operation system is classified as a normal operation state.

8. A mining equipment control method according to claim 7, characterized in that: When an abnormal operation state is generated during the operation of the automatic operation system, a comprehensive analysis is performed on several amplitude fluctuation indexes generated by the subsequent operation of the automatic operation system and the preset amplitude fluctuation index reference threshold to generate the operation evaluation coefficient YX XS , based on the formula: Where AF r represents the amplitude fluctuation index greater than or equal to the amplitude fluctuation index reference threshold, r represents the number of the amplitude fluctuation index greater than or equal to the amplitude fluctuation index reference threshold, r=1, 2, 3, ..., e, e is a positive integer; The generated evaluation coefficient is compared and analyzed with the preset evaluation coefficient reference threshold, and the analysis results are as follows: If the evaluation coefficient is greater than or equal to a preset evaluation coefficient reference threshold, the abnormal operation state of the automated operation system is further classified as a non-accidental abnormality; If the evaluation coefficient is less than a preset evaluation coefficient reference threshold, the abnormal operation state of the automation operation system is further classified as an accidental abnormality.

9. A mining equipment control system, used to implement a mining equipment control method as described in any one of claims 1 to 8, characterized in that: It includes parameter real-time monitoring module, abnormal analysis and fluctuation assessment module, operation status identification module, abnormal classification module and intelligent response and processing module: The real-time parameter monitoring module obtains the various parameter information generated during the crushing process of the crusher in real time during the operation of the crusher, including vibration response characteristics and energy transfer and dissipation characteristics; The abnormal analysis and fluctuation assessment module performs abnormal analysis on the acquired vibration response characteristics and energy transfer and dissipation characteristics, and inputs the vibration response characteristics and energy transfer and dissipation characteristics after abnormal analysis into the pre-learned machine learning model, generates the amplitude fluctuation index through the machine learning model, and performs intelligent evaluation on the crusher operation process through the amplitude fluctuation index; The operation status recognition module divides the operation process of the automated operation system into normal operation state and abnormal operation state based on the evaluation results of the machine learning model, and intelligently perceives the abnormal crushing process of the crusher through the divided operation states; The abnormality classification module further classifies the abnormal operation status of the automation system into accidental abnormality or non-accidental abnormality based on the abnormal operation status evaluated by the machine learning model; The intelligent response and processing module takes different measures for different anomalies.

Images