Mine hoist intelligent control method and system
By constructing a speed-adaptive penalty factor and dynamic benchmark potential energy, combined with local outlier factor analysis, the problems of false alarms and speed condition interference caused by stiffness changes in deep well operation of mine hoists were solved, achieving accurate monitoring and fault identification throughout the entire stroke.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- LUOYANG DIANJING INTELLIGENT CONTROL TECH CO LTD
- Filing Date
- 2026-03-13
- Publication Date
- 2026-05-15
AI Technical Summary
In deep well operations, mine hoists suffer from false alarms and nonlinear interference in speed conditions due to changes in system stiffness with depth. Existing monitoring technologies cannot achieve accurate monitoring across the entire speed range.
By constructing a speed-adaptive penalty factor and a dynamic benchmark potential energy, combined with local outlier factor analysis, the monitoring threshold is dynamically adjusted to achieve intelligent control of the mine hoist, eliminate false alarms in deep wells, and improve low-speed sensitivity.
It achieves accurate monitoring throughout the entire process, significantly improves the sensitivity to minor faults, ensures the stability and anti-interference ability of the system under different operating conditions, and avoids false alarms and missed alarms.
Smart Images
Figure CN121823348B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of mining machinery control technology. More specifically, this invention relates to an intelligent control method and system for mine hoists. Background Technology
[0002] Mine hoists, as a key hub in the mining production system, undertake the important tasks of hoisting coal and ore, as well as transporting personnel and equipment. In the operation of mines thousands of meters deep, the hoisting container primarily relies on guideways for guidance. However, due to the extremely harsh environment inside the shaft, factors such as persistent dampness, corrosion, and geological pressure often lead to problems such as misalignment, wear, or deformation of the guideways. Once a guideway malfunctions, it can easily trigger severe mechanical impacts, and in serious cases, even lead to catastrophic safety accidents such as canister jamming or rope breakage.
[0003] Current monitoring technologies face two major technical challenges in practical applications.
[0004] Firstly, there's the issue of false alarms in deep wells due to the coupling effect between depth and stiffness. As an elastic body, the length of the hoisting wire rope increases with the depth of the hoisting container, causing the lateral stiffness of the hoisting system to gradually decrease from its maximum value at the wellhead. In deep well areas, even if the guideway is in normal condition, the container will experience significant natural swaying due to the reduced system stiffness. Existing technologies typically use fixed alarm thresholds, such as three times the acceleration due to gravity. This static standard is highly susceptible to false alarms in deep wells. To avoid frequent false alarms, field personnel are often forced to increase the overall alarm threshold, but this ultimately leads to real faults in shallow wells being missed due to insufficient sensitivity.
[0005] Secondly, there is the issue of nonlinear interference under speed conditions. Mine hoists operate at extremely wide speeds, typically ranging from zero to twelve meters per second. During high-speed operation, the significant aerodynamic and mechanical noise often masks subtle fault characteristics at lower speeds. Furthermore, the vibration energy generated by the same track unevenness varies greatly at different speeds, resulting in a lack of a unified evaluation standard and hindering accurate monitoring across the entire speed range. Summary of the Invention
[0006] The purpose of this invention is to propose an intelligent control method and system for mine hoists to solve the problem of false alarms in deep wells caused by the change in system stiffness with depth in the prior art; to this end, this invention provides solutions in the following two aspects.
[0007] In a first aspect, the present invention provides an intelligent control method for a mine hoist, comprising:
[0008] The system acquires the lateral vibration acceleration signal of the hoisting container, the real-time depth of the hoist, and the real-time speed of the hoist. A speed-adaptive penalty factor is constructed based on the real-time speed, and variational mode decomposition is performed on the lateral vibration acceleration signal using this factor to obtain high-frequency components. Based on the real-time depth and speed of the hoist, combined with the allowable static noise floor, the rated maximum kinetic energy vibration increment, and the reference depth, a dynamic reference potential energy is constructed that varies with the operating conditions. This dynamic reference potential energy is positively correlated with the square of the real-time speed and the logarithm of the real-time depth. The envelope amplitude of the high-frequency components is extracted as the observed value, and the relative overscaling between the observed value and the dynamic reference potential energy is calculated. Simultaneously, the difference mean of the high-frequency components is calculated, and an inverse velocity weighting term is calculated based on the real-time speed and the velocity deviation constant. The impact risk index is determined based on the sum of the relative overscaling and the inverse velocity weighting term. The impact risk index is input into a sliding time window for local outlier analysis. If the impact risk index exceeds a preset warning threshold and the duration meets preset conditions, a fault is identified, and a control signal is output, thereby achieving intelligent control of the mine hoist.
[0009] Thus, by constructing a dynamic benchmark potential energy that varies with depth and speed, it can automatically adapt to the natural swaying caused by reduced stiffness in deep wells, eliminating false alarms in deep wells from a physical perspective. At the same time, by calculating the impact risk index inverse velocity weighting, it can amplify the sensitivity to minor faults during low-speed operation, achieving accurate monitoring across the entire speed range.
[0010] Preferably, the step of constructing the speed adaptive penalty factor based on the real-time speed includes: calculating the speed adaptive penalty factor at the current moment using the following formula: In the formula, This represents the penalty factor of the variational mode decomposition algorithm at the current moment, and is a dimensionless value; This represents the low-speed reference penalty factor, which is a dimensionless value. This indicates the rated maximum speed of the hoist, in m / s; This indicates real-time speed, expressed in m / s. This represents the bandwidth sensitivity coefficient, which is a dimensionless value.
[0011] Thus, by introducing a speed-related adaptive penalty factor, a larger penalty factor is used at low speeds to obtain high frequency resolution, and the penalty factor is automatically reduced at high speeds to broaden the modal bandwidth, preventing the loss of high-frequency broadband signals caused by high-speed impacts and ensuring the effectiveness of signal decomposition across the entire speed range.
[0012] Preferably, the construction of the dynamic reference potential energy that varies with operating conditions includes: calculating the dynamic reference potential energy using the following relationship: In the formula, This represents the dynamic baseline potential energy, with units of gravitational acceleration g. This represents the allowable static noise floor value, in units of gravitational acceleration g. This represents the rated maximum kinetic energy vibration increment, expressed in gravitational acceleration g. This indicates the rated maximum speed of the hoist, in m / s; This indicates real-time speed, expressed in m / s. This indicates the real-time depth, in meters (m). Indicates the reference depth, in meters (m). Represents the natural constant.
[0013] Thus, by incorporating the kinetic energy theorem (the square term of velocity) and the stiffness characteristics of the cantilever beam (the logarithmic term of depth) into the calculation of the benchmark value, the monitoring threshold can be dynamically adjusted according to the physical conditions. This allows the system to accommodate normal large-amplitude swaying in deep wells and accurately capture abnormal vibrations in shallow wells, significantly improving the system's environmental adaptability.
[0014] Preferably, determining the impact risk index includes: calculating the impact risk index using the following formula: In the formula, This represents the impact risk index, which is a dimensionless value. The envelope amplitude represents the high-frequency component, expressed in gravitational acceleration g. This represents the dynamic baseline potential energy, with units of gravitational acceleration g. This represents the larger of the zero-sum and the difference between the observed value and the dynamic baseline potential energy; This represents the weighting coefficient, which is a dimensionless value. Indicates the sampling window length, which is a positive integer; This represents the difference in the signal within the window, expressed in gravitational acceleration g. This represents the allowable static noise floor value, in units of gravitational acceleration g. This indicates real-time speed, expressed in m / s. This represents the velocity deviation constant, with units of m / s; This indicates the rated maximum speed of the hoist, expressed in m / s.
[0015] Thus, by making the difference term and the velocity term dimensionless, the risk of mathematical singularity caused by the inconsistency of dimensions is eliminated; at the same time, by introducing inverse velocity weighted logic, the sensitivity to signal changes at low speeds is greatly improved, thereby giving the system a microscope effect in low-speed rope inspection or maintenance mode, which can effectively detect hidden dangers such as misalignment of small joints in the tank passage.
[0016] Preferably, the calculation of the differential mean of the high-frequency components specifically includes: acquiring the high-frequency component data sequence within the sampling window; performing a first-order difference calculation on the data sequence to obtain an acceleration sequence, which reflects the rate of change of the vibration signal; and calculating the average value of the absolute value of the acceleration sequence as the differential mean to eliminate the influence of the sampling window length variation on the numerical fluctuation.
[0017] Preferably, the specific process of the local outlier analysis includes: establishing a sliding time window containing historical impact risk indices; calculating the local reachability density of the impact risk index at the current moment relative to historical data within the sliding time window; calculating the local outlier value based on the local reachability density to assess the degree of anomaly in the current data, so as to distinguish between transient disturbances and persistent mechanical failures.
[0018] Thus, by establishing a sliding time window that incorporates historical impact risk indices and calculating the local reachability density of the current data relative to historical data, a local outlier factor value is derived to assess the degree of anomaly in the current data. This historical context-based analysis mechanism avoids the shortcomings of triggering alarms solely based on absolute values at a single moment. It enables the algorithm to accurately identify and automatically filter brief and harmless sudden fluctuations, ensuring that the system ultimately outputs control signals only for genuine mechanical impact faults. This significantly improves the accuracy of intelligent decision-making and the overall anti-interference capability of the system.
[0019] Preferably, the specific logic for determining a fault is as follows: setting a warning threshold and a duration threshold; when the calculated impact risk index is less than the warning threshold, the system is determined to be in a safe operating state; when the calculated impact risk index is greater than or equal to the warning threshold, and the duration of this state exceeds the duration threshold, the system is determined to have experienced a mechanical impact fault.
[0020] Preferably, the static noise floor allowable value is used to ensure that the dynamic reference potential energy is not zero when the real-time velocity is zero, so as to avoid the denominator being zero in subsequent calculations and to cover the electronic noise floor of the sensor itself.
[0021] Preferably, the bandwidth sensitivity coefficient is used to adjust the sensitivity of the penalty factor to changes in speed. When the real-time speed increases, the penalty factor decreases to broaden the modal bandwidth, thereby capturing wideband signals under high-speed impact.
[0022] In a second aspect, an intelligent control system for a mine hoist includes:
[0023] The system includes a processor and a memory storing computer instructions for intelligent control of a mine hoist, which, when executed by the processor, cause the system to perform the aforementioned intelligent control method for the mine hoist.
[0024] The beneficial effects of this invention are as follows: This invention can automatically adapt to the natural swaying of the hoisting container caused by reduced stiffness in deep wells, effectively eliminating false alarms due to increased depth and achieving accurate monitoring throughout the entire process. In low-speed operation mode, this solution significantly improves the sensitivity to minor fault characteristics, effectively identifying subtle hidden dangers such as misalignment of the hoisting joints, achieving the dual goals of low-speed inspection for potential hazards and high-speed operation for safety. Furthermore, this invention ensures the stability of the algorithm output under different operating conditions such as stationary, startup, and full-speed operation, avoiding the risk of mathematical calculation divergence, thereby significantly improving the environmental adaptability and operational reliability of the mine hoist intelligent control system. Attached Figure Description
[0025] Figure 1 This is a flowchart of an intelligent control method for a mine hoist according to an embodiment of the present invention;
[0026] Figure 2 This is a comparison diagram of vibration characteristic distribution based on depth and stiffness coupling characteristics according to an embodiment of the present invention;
[0027] Figure 3 This is a cluster distribution diagram of the operating state after introducing speed correction according to an embodiment of the present invention. Detailed Implementation
[0028] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
[0029] like Figure 1 As shown in this embodiment, an intelligent control method for a mine hoist includes the following steps:
[0030] Step S1: Obtain the lateral vibration acceleration signal of the hoisting container, the real-time depth of the hoist, and the real-time speed of the hoist. Construct a speed adaptive penalty factor based on the real-time speed. Use the speed adaptive penalty factor to perform variational mode decomposition on the lateral vibration acceleration signal to obtain high-frequency components.
[0031] Specifically, this step aims to complete signal acquisition and preprocessing. In terms of data acquisition, the system utilizes piezoelectric accelerometers installed on the side wall of the lifting container to collect lateral vibration acceleration signals in real time. The sampling rate is set to 1000Hz to ensure accurate capture of transient impacts; simultaneously, the real-time depth of the hoisting container is synchronously acquired via the hoist's PLC main control system or a rotary encoder on the drum shaft. (Unit: m, with the wellhead as 0 and downward as positive) and real-time velocity (Unit: m / s).
[0032] To extract features characterizing guide faults from broadband mechanical noise, a variational mode decomposition (VMD) algorithm is employed. Addressing the issue of fault frequency drift caused by hoist speed variations, this invention abandons the fixed-parameter approach and constructs a speed-adaptive penalty factor. .
[0033] The calculation formula is as follows:
[0034] ;
[0035] In the formula, This represents the penalty factor of the variational mode decomposition algorithm at the current moment, and is a dimensionless value; This represents the low-speed reference penalty factor, which is a dimensionless value and takes the value of 2500. At this value, VMD has a high frequency resolution. This indicates the rated maximum speed of the hoist, in m / s, with a value of 12 m / s. This indicates real-time speed, expressed in m / s. This represents the bandwidth sensitivity coefficient, a dimensionless value with a value of 1.5.
[0036] Step S2: Based on the real-time depth and real-time speed of the hoist, combined with the allowable static noise level, the rated maximum kinetic energy vibration increment, and the reference depth, a dynamic reference potential energy that varies with the working conditions is constructed. The dynamic reference potential energy is positively correlated with the square of the real-time speed and positively correlated with the logarithm of the real-time depth.
[0037] Specifically, this step aims to calculate a "dynamic threshold" that varies with operating conditions, i.e., a dynamic reference potential energy. (Unit: g) This indicator represents "the upper limit of the reasonable vibration amplitude allowed by the system at the current depth and velocity".
[0038] The calculation formula is as follows:
[0039] ;
[0040] In the formula, This represents the dynamic baseline potential energy, with units of gravitational acceleration g. This represents the allowable static noise floor value, expressed in gravitational acceleration g, with a value of 0.2g. This represents the rated maximum kinetic energy vibration increment, expressed in gravitational acceleration g, with a value of 1.5g. This indicates the rated maximum speed of the hoist, in m / s; This indicates real-time speed, expressed in m / s. This indicates the real-time depth, in meters (m). This represents a reference depth in meters (m), and is taken as 10% of the total wellbore depth. Represents the natural constant.
[0041] The above expression for dynamic benchmark potential energy constructs a dynamic benchmark potential energy evaluation mechanism based on the kinetic energy theorem and the stiffness characteristics of cantilever beams. It aims to solve the contradiction between false alarms and false alarms in complex deep well conditions under traditional fixed thresholds. Its core logic is to map the real-time operating state of the lifting system to the physically permissible upper limit of vibration, and to dynamically adjust the alarm benchmark in real time through the nonlinear coupling of velocity and depth.
[0042] When the hoist is in the shallow well area or stationary, the system's lateral stiffness is extremely high due to the short length of the wire rope, and the operating speed approaches zero. At this time, the kinetic energy term approaches zero due to the extremely low speed, and the depth logarithm term in the formula reverts to unit 1 due to the introduction of the natural constant. Therefore, the calculated dynamic baseline potential energy is mainly determined by the allowable static noise floor. The algorithm thus exhibits high sensitivity, maintaining the baseline value at the lowest level covering sensor electronic noise and equipment micro-vibrations. This allows for accurate detection of minor mechanical anomalies that may occur at this time, preventing missed detection of shallow well faults due to excessively high thresholds.
[0043] When the hoisting container moves into the deep well area or enters the high-speed operation phase, the lateral stiffness of the system decreases significantly with the increase of the wire rope length, exhibiting "soft stiffness" characteristics. Simultaneously, high-speed operation is accompanied by a huge surge in kinetic energy. At this point, the logarithmic term of depth increases with increasing depth, while the quadratic term of velocity increases parabolically with increasing kinetic energy. The coupling effect of these two factors causes the dynamic reference potential energy to rise rapidly. This change forces the algorithm to automatically "tolerate" the natural large swaying and high-speed aerodynamic noise caused by the decrease in stiffness at deep wells, greatly improving the tolerance to large vibrations under normal operating conditions. Therefore, from a physical perspective, this avoids the problem of false alarms in deep wells caused by "depth-stiffness" coupling.
[0044] Step S3: Extract the envelope amplitude of the high-frequency component as the observation value, calculate the relative overscaling of the observation value and the dynamic reference potential energy, and calculate the difference mean of the high-frequency component. Combine the real-time velocity and velocity deviation constant to calculate the inverse velocity weighting term. Determine the impact risk index based on the sum of the relative overscaling and the inverse velocity weighting term.
[0045] Specifically, the envelope amplitude of the high-frequency components decomposed in step S1 is extracted as the observation value. Calculate the final judgment criterion—the impact risk index. .
[0046] The impact risk index is calculated using the following formula:
[0047] ;
[0048] In the formula, This represents the impact risk index, which is a dimensionless value. The envelope amplitude represents the high-frequency component, expressed in gravitational acceleration g. This represents the dynamic baseline potential energy, with units of gravitational acceleration g. This represents the larger of the zero-sum and the difference between the observed value and the dynamic baseline potential energy; This represents the weighting coefficient, a dimensionless value with a value of 5.0. Indicates the sampling window length, which is a positive integer; This indicates the sequence number of each data point within the current sliding sampling window; This represents the difference in the signal within the window, expressed in gravitational acceleration g. This represents the allowable static noise floor value, in units of gravitational acceleration g. This indicates real-time speed, expressed in m / s. This represents the velocity deviation constant, with units of m / s and a value of 0.5 m / s, used to prevent the denominator from being 0. This indicates the rated maximum speed of the hoist, expressed in m / s.
[0049] The above-mentioned formula for calculating the impact risk index constructs a dimensionless impact risk index calculation mechanism that integrates relative overscaling and inverse velocity weighting terms. Its core logic lies in strictly normalizing the envelope amplitude characteristics of high-frequency components and the differential characteristics of the signal within the window, and dynamically adjusting the contribution of the differential characteristics to the total risk value according to the real-time operating status, thereby stably outputting the judgment index used to finally determine the mechanical impact.
[0050] In calculating the first part of the relative overscaling, the formula extracts the envelope amplitude of the high-frequency components as the observed value and preliminarily assesses the current vibration state by calculating the difference between the observed value and the dynamic reference potential energy. The numerator of the formula uses the logic of taking zero and the larger of the difference between the observed value and the dynamic reference potential energy, forcing the results corresponding to safe observed values that have not exceeded the dynamic reference potential energy boundary to be zero, ensuring that the system only accumulates risk for abnormal vibrations that substantially exceed the limit. The calculation result is then divided by the dynamic reference potential energy and a square root operation is performed. This not only allows the relative overscaling to smoothly reflect the severity of the observed value exceeding the limit, but also eliminates the physical unit of gravitational acceleration at the mathematical level, realizing the normalization of the evaluation index.
[0051] In calculating the inverse velocity weighting term in the second part, the formula performs dimensionless deep processing on both the difference sequence and the velocity variable. The formula first calculates the average of the absolute values of the signal differences within the window and divides it by the allowable static noise floor. This operation transforms the absolute rate of physical change into a relative abrupt change coefficient relative to the underlying environmental background noise. Subsequently, the formula divides the sum of the real-time velocity and the velocity deviation constant by the hoist's rated maximum speed to construct a normalized operating speed proportional denominator. When the hoist is operating at high speed, this speed proportional denominator is at a high value, effectively suppressing the overall calculated value of the inverse velocity weighting term. This significantly reduces the interference of normal broadband mechanical fluctuations under high-speed operating conditions on the final impact risk index, ensuring the accuracy of safety judgments during high-speed operation.
[0052] When the hoist is in low-speed inspection mode, the normalized speed ratio denominator decreases sharply, while the value of the inverse speed weighting term, combined with the weighting coefficient, increases exponentially, thus amplifying the contribution of the high-frequency component differential characteristics to the impact risk index. This inverse weighting mechanism makes the system extremely sensitive to minor mechanical impacts at low speeds, giving it the ability to detect early potential hazards with high sensitivity. Simultaneously, the introduction of the speed deviation constant in the denominator prevents computational crashes caused by the denominator becoming zero when the real-time speed drops to zero. Furthermore, the division of each variable by a reference value of the same unit ensures that the values on both sides of the plus sign are physically unified as dimensionless pure numbers, guaranteeing the rigor of the overall mathematical model and the absolute stability of the system output.
[0053] Step S4: Input the impact risk index into the sliding time window for local outlier factor analysis. If the impact risk index is greater than the preset warning threshold and the duration meets the preset conditions, it is determined to be a fault and a control signal is output, thereby realizing intelligent control of the mine hoist.
[0054] Specifically, the calculated Input the data into a sliding time window for LOF (Local Outlier) analysis. Set an alert threshold. .
[0055] Next, we will perform a logical check:
[0056] like It was determined to be safe;
[0057] like And duration This is determined to be a fault.
[0058] The duration determination here is to filter out single-point spikes caused by electromagnetic interference and only respond to continuous impacts with mechanical entities.
[0059] In this way, by performing outlier analysis on the risk index after it has been "purified" by the physical model, the actual fault state can be accurately identified, achieving precise classification control with zero false alarms and zero false alarms.
[0060] To more intuitively illustrate the beneficial effects of the present invention, please refer to the appendix to the specification below. Figure 2 and attached Figure 3 Further explanation is needed.
[0061] Figure 2 This figure illustrates a comparison of vibration characteristic distributions based on the coupling characteristics of depth and stiffness in embodiments of the present invention. The horizontal axis represents the depth of the lifting container, and the vertical axis represents the amplitude of the lateral vibration characteristics. The dashed horizontal line represents a fixed alarm line commonly used in the prior art, whose setpoint remains constant and does not change with depth. The thick solid line rising non-linearly from left to right represents the dynamic reference potential energy boundary constructed in this invention. The dense dots, whose distribution range gradually expands with increasing depth, represent normal operation monitoring points, while the pentagrams scattered in areas with relatively high amplitudes represent actual guide failure points.
[0062] from Figure 2 As can be seen, during the shallow well operation phase, due to the shorter wire rope and higher system stiffness, the vibration amplitude at normal operation monitoring points is generally low. The dynamic reference potential energy boundary of this invention is correspondingly maintained at a low level, thus enabling it to sensitively detect relatively small abnormal impacts in this area. With the continuous increase in lifting depth, the lateral stiffness of the lifting system gradually decreases due to the longer wire rope, and the vibration amplitude range of the normal operation monitoring points significantly expands, exhibiting a physical phenomenon of intensified natural swaying. Especially in the deep well operation zone, the vibration amplitude of a large number of normal operation monitoring points naturally increases. At this point, although these normal values are far from posing a mechanical impact hazard, they have significantly exceeded the horizontal straight line, forming a clear false alarm zone in existing technologies. This phenomenon makes traditional monitoring systems prone to triggering false alarms in deep wells. In contrast, the dynamic reference potential energy boundary constructed by this invention rises synchronously with increasing depth based on a physical coupling mechanism, closely adhering to and perfectly enveloping the upper edge of all normal operation monitoring points, automatically tolerating normal large-amplitude swaying in the deep well region. Meanwhile, all the five-pointed stars are clearly suspended above the thick solid line, which enables the system to accurately identify real mechanical impact faults across the entire depth range. This intuitively demonstrates that the present invention effectively eliminates the false alarm problem of the fixed threshold strategy under complex deep well conditions, and achieves accurate classification and dynamic defense of operational risks throughout the entire process.
[0063] Figure 3This diagram illustrates the clustering distribution of operating states after speed correction is introduced in this embodiment of the invention. The horizontal axis represents the increased real-time operating speed of the container, and the vertical axis represents the final impact risk index output by the algorithm. The horizontal dotted line at the risk index of 4 represents the system's preset intelligent decision boundary, which clearly divides the entire chart space into a safe operating zone below and a high-risk decision zone above. The densely clustered circular markers at the bottom of the chart represent safe operating state samples, while the scattered diamond-shaped markers in the upper half of the chart represent high-risk impact samples.
[0064] from Figure 3 As can be seen, across the entire speed range from standstill to maximum speed, thanks to the introduction of the inverse velocity weighting term in the core formula of this invention, the algorithm effectively suppresses the nonlinear interference caused by normal aerodynamic and mechanical noise in the high-speed segment. This results in the calculations under massive normal operating conditions being smoothly and stably compressed into a very low safe value range below the boundary line, exhibiting extremely excellent clustering characteristics. Meanwhile, once the hoisting system experiences a real mechanical impact, regardless of whether it is in a low-speed or high-speed hoisting state, the impact risk index, after being purified by both relative overscaling and inverse velocity weighting mechanisms, will be significantly amplified. At this point, the diamond-shaped marker representing the fault floats high above the intelligent decision boundary line, forming a huge numerical difference with the normal samples at the bottom. Based on this, the system can keenly and accurately judge mechanical impact faults, intuitively demonstrating that this invention overcomes the difficulty of fault feature extraction caused by drastic changes in hoist speed, achieving accurate separation and intelligent decision-making between normal fluctuations and real faults across the entire speed range.
[0065] The present invention also provides an intelligent control system for a mine hoist. The system includes a processor and a memory, the memory storing computer program instructions. When the processor executes the computer program instructions, it implements the intelligent control method for a mine hoist according to the present invention described above.
[0066] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and therefore will not be described in detail here.
[0067] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented by computer-readable / executable instructions stored or otherwise maintained on such a computer-readable medium.
[0068] In the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.
[0069] While various embodiments of the invention have been shown and described in this specification, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention.
Claims
1. A method for intelligent control of a mine hoist, characterized in that, include: The lateral vibration acceleration signal of the hoisting container, the real-time depth of the hoist, and the real-time speed of the hoist are acquired. A speed adaptive penalty factor is constructed based on the real-time speed. The lateral vibration acceleration signal is then subjected to variational mode decomposition using the speed adaptive penalty factor to obtain the high-frequency components. Based on the real-time depth and real-time speed of the hoist, combined with the allowable static noise floor value, the rated maximum kinetic energy vibration increment and the reference depth, a dynamic reference potential energy that varies with the working conditions is constructed. The dynamic reference potential energy is positively correlated with the square of the real-time speed and positively correlated with the logarithm of the real-time depth. The envelope amplitude of the high-frequency component is extracted as the observation value. The relative overscaling of the observation value and the dynamic reference potential energy is calculated. At the same time, the difference mean of the high-frequency component is calculated. The inverse velocity weighting term is calculated in combination with the real-time velocity and the velocity deviation constant. The impact risk index is determined based on the sum of the relative overscaling and the inverse velocity weighting term. The impact risk index is input into a sliding time window for local outlier factor analysis. If the impact risk index is greater than the preset warning threshold and the duration meets the preset conditions, it is determined to be a fault and a control signal is output, thereby realizing intelligent control of the mine hoist.
2. The intelligent control method for mine hoists according to claim 1, characterized in that, The step of constructing a speed adaptive penalty factor based on real-time speed includes: The adaptive velocity penalty factor at the current moment is calculated using the following formula: ； In the formula, This represents the penalty factor of the variational mode decomposition algorithm at the current moment, and is a dimensionless value; This represents the low-speed reference penalty factor, which is a dimensionless value. This indicates the rated maximum speed of the hoist, in m / s; This indicates real-time speed, expressed in m / s. This represents the bandwidth sensitivity coefficient, which is a dimensionless value.
3. The intelligent control method for mine hoists according to claim 1, characterized in that, The construction of the dynamic benchmark potential energy that varies with operating conditions includes: The dynamic baseline potential energy is calculated using the following formula: ； In the formula, This represents the dynamic baseline potential energy, with units of gravitational acceleration g. This represents the allowable static noise floor value, in units of gravitational acceleration g. This represents the rated maximum kinetic energy vibration increment, expressed in gravitational acceleration g. This indicates the rated maximum speed of the hoist, in m / s; This indicates real-time speed, expressed in m / s. This indicates the real-time depth, in meters (m). Indicates the reference depth, in meters (m). Represents the natural constant.
4. The intelligent control method for mine hoists according to claim 1, characterized in that, The determination of the impact risk index includes: The impact risk index is calculated using the following formula: ； In the formula, This represents the impact risk index, which is a dimensionless value. The envelope amplitude represents the high-frequency component, expressed in gravitational acceleration g. This represents the dynamic baseline potential energy, with units of gravitational acceleration g. This represents the larger of the zero-sum and the difference between the observed value and the dynamic baseline potential energy; This represents the weighting coefficient, which is a dimensionless value. Indicates the sampling window length, which is a positive integer; This represents the difference in the signal within the window, expressed in gravitational acceleration g. This represents the allowable static noise floor value, in units of gravitational acceleration g. This indicates real-time speed, expressed in m / s. This represents the velocity deviation constant, with units of m / s; This indicates the rated maximum speed of the hoist, expressed in m / s.
5. The intelligent control method for mine hoists according to claim 4, characterized in that, The calculation of the difference mean of the high-frequency components specifically includes: Acquire the high-frequency component data sequence within the sampling window; The data sequence is subjected to first-order difference calculation to obtain the jerk sequence, which reflects the rate of change of the vibration signal; The average of the absolute values of the accelerometer sequence is calculated as the difference mean to eliminate the influence of the sampling window length variation on numerical fluctuations.
6. The intelligent control method for mine hoists according to claim 1, characterized in that, The specific process of the local outlier factor analysis includes: Establish a sliding time window that includes historical impact risk indices; Calculate the local reachability density of the impact risk index at the current moment relative to historical data within the sliding time window; The local outlier factor value is calculated based on the local reachability density to assess the degree of anomaly in the current data, in order to distinguish between transient disturbances and persistent mechanical failures.
7. The intelligent control method for mine hoists according to claim 1, characterized in that, The specific logic for determining a fault is as follows: Set warning thresholds and duration thresholds; When the calculated impact risk index is less than the warning threshold, the system is determined to be in a safe operating state. When the calculated impact risk index is greater than or equal to the warning threshold, and the duration of this state exceeds the duration threshold, the system is determined to have experienced a mechanical impact failure.
8. The intelligent control method for mine hoists according to claim 3, characterized in that, The static noise floor allowance is used to ensure that the dynamic reference potential energy is not zero when the real-time velocity is zero, so as to avoid the denominator being zero in subsequent calculations and to cover the electronic noise floor of the sensor itself.
9. The intelligent control method for mine hoists according to claim 2, characterized in that, The bandwidth sensitivity coefficient is used to adjust the sensitivity of the penalty factor to changes in speed. When the real-time speed increases, the penalty factor decreases to widen the modal bandwidth, thereby capturing wideband signals under high-speed impact.
10. An intelligent control system for a mine hoist, characterized in that, include: processor; A memory storing computer instructions for intelligent control of a mine hoist, which, when executed by the processor, cause the system to perform the intelligent control method for a mine hoist according to any one of claims 1-9.