Medium-speed coal mill vibration early warning method based on material level state of pebble coal bunker

Abstract

The invention relates to a vibration early-warning method for a medium-speed coal mill based on the material level state of a pebble coal bunker, and belongs to the technical field of coal mill monitoring. The vibration early-warning method comprises the following steps: obtaining material level data and vibration data of the medium-speed coal mill, and calculating a material level fluctuation rate based on the material level data; inputting the vibration data into a fast Fourier transform (FFT) algorithm to obtain a complex frequency spectrum vector, and further obtaining a dominant vibration amplitude and a vibration energy entropy; and a material level vibration correlation model is constructed according to the material level fluctuation rate and the material level data, the vibration amplitude of the pebble coal bunker is output, the deviation between the vibration amplitude and the dominant vibration amplitude is calculated, and a coupling early warning index is obtained in combination with the vibration energy entropy. And calculating the mean value of the coupling early warning indexes, and determining an early warning threshold value by combining the material level fluctuation ratio. And dividing risk levels according to the coupling early warning index and the early warning threshold. The FFT algorithm is adopted to extract the dominant vibration amplitude and the energy entropy, a dynamic correlation model is constructed in combination with the material level fluctuation rate, and the logic mapping relation between the material level change and the vibration response is established.

Description

TECHNICAL FIELD

[0001] The present application relates to a medium-speed coal mill vibration early warning method based on the stone coal bunker level state, belonging to the technical field of coal mill monitoring. BACKGROUND

[0002] Medium-speed coal mill is the core equipment in modern thermal power plants, chemical plants and other industrial production processes, and its main function is to grind raw coal into coal powder that meets the requirements of combustion or process. Its stable operation has a decisive influence on the efficiency, safety and economy of the entire production process. However, in the actual operation process, medium-speed coal mills often face vibration problems, which not only affect the service life of the equipment, but also may cause serious mechanical failures, even lead to shutdown for maintenance, causing huge economic losses to enterprises. The sources of vibration problems are various, including imbalance of grinding roller and grinding disc, accumulation of stone coal (coal gangue, impurities, etc. that are not completely ground), wear of lining plate, uneven coal supply and loosening of internal parts of the equipment, etc. Especially, the change of stone coal bunker level has a significant influence on the vibration characteristics of the coal mill. As the terminal equipment of the coal mill slagging system, the level state of the stone coal bunker directly reflects the discharge of stone coal inside the coal mill. When the stone coal bunker level appears abnormal fluctuation, it will cause unbalanced stress inside the coal mill, and then cause the increase of equipment vibration amplitude. If not monitored and warned in time, this abnormal vibration may accelerate the wear of equipment parts, and even cause more serious failures, such as grinding roller fracture, bearing damage, etc.

[0003] At present, vibration monitoring technology has been applied in the industrial field to some extent, but most methods only focus on the vibration data itself, ignoring the influence of other key factors (such as the level state of stone coal bunker) in the coal mill system on the vibration characteristics. In addition, traditional vibration monitoring technology still has obvious deficiencies in feature extraction, dynamic early warning and intelligent risk assessment. SUMMARY

[0004] In order to solve the problems existing in the prior art, the present application proposes a medium-speed coal mill vibration early warning method based on the level state of stone coal bunker.

[0005] The technical scheme of the present application is as follows:

[0006] A medium-speed coal mill vibration early warning method based on the level state of stone coal bunker, comprising the following steps:

[0007] Obtaining the level data and vibration data of the medium-speed coal mill, and obtaining the level fluctuation rate based on the level data;

[0008] Taking the vibration data as the input of the fast Fourier transform algorithm FFT, outputting a complex spectrum vector, and obtaining the dominant vibration amplitude and vibration energy entropy based on the complex spectrum vector;

[0009] constructing a stock level vibration correlation model based on the stock level fluctuation rate and the stock level data, and outputting a vibration amplitude of the stone coal bunker;

[0010] calculating a deviation between the dominant vibration amplitude and the vibration amplitude of the stone coal bunker, and obtaining a coupling early warning index based on the deviation and a vibration energy entropy;

[0011] calculating a mean value of the coupling early warning index, and obtaining a warning threshold based on the mean value and the stock level fluctuation rate;

[0012] obtaining a risk level based on the coupling early warning index and the warning threshold.

[0013] Preferably, the method further comprises filtering the stock level data and the vibration data;

[0014] The filtered stock level data is expressed by the following formula:

[0015] L f (t)=α L ·L raw (t)+(1-α L )·L f (t-1);

[0016] In the formula, L f (t) represents the filtered stock level data at time t, α L represents a first-order low-pass filtering coefficient of the stock level data, L raw (t) represents the stock level data at time t, and L f (t-1) represents the filtered stock level data at time t-1.

[0017] The filtered vibration data is expressed by the following formula:

[0018] V f (t)=α V ·V raw (t)+(1-α V )·V f (t-1);

[0019] In the formula, V f (t) represents the filtered vibration data at time t, α V represents a first-order low-pass filtering coefficient of the vibration data, V raw (t) represents the vibration data at time t, and V f (t-1) represents the filtered vibration data at time t-1.

[0020] Preferably, the stock level fluctuation rate is expressed by the following formula:

[0021]

[0022] wherein R w (t) represents the fluctuation rate of the material level at time t, ΔT represents the length of the time window of the material level data, N represents the number of sampling points of the length of the time window, Δt represents the data sampling time interval, L f (i) represents the filtered material level data at time i, L f (i-1) represents the filtered material level data at time i-1.

[0023] Preferably, the vibration data is taken as the input of the fast Fourier transform algorithm FFT, and the specific steps are as follows:

[0024] A vibration window vector is constructed by taking N w filtered vibration data at past time points from time t, which is expressed by the formula as follows:

[0025]

[0026] wherein V window (t) represents the vibration window vector at time t, V f (t-N w +2) represents the filtered vibration data at time t-N w +2;

[0027] The vibration window vector is windowed, which is expressed by the formula as follows:

[0028] x(t) = V window (t)⊙w;

[0029] wherein ⊙ represents the Hadamard product operator, w represents the window function vector, and x(t) represents the windowed vibration window vector at time t;

[0030] The windowed vibration window vector is taken as the input of the fast Fourier transform algorithm FFT, and the complex spectrum vector is output, which is expressed by the formula as follows:

[0031] X(t) = FFT(x(t));

[0032] wherein X(t) represents the complex spectrum vector at time t, and FFT represents the fast Fourier transform algorithm function;

[0033] The complex spectrum vector includes N w complex spectrum values, which is expressed by the formula as follows:

[0034] X(t) = [X(1,t), …, X(m,t), …, X(N w ,t)] T ;

[0035] X(m, t) represents a complex spectrum value of the mth frequency band of the complex spectrum vector at the t time point;

[0036] Taking the maximum complex spectrum value of the complex spectrum vector in the preset frequency band interval [M min ,M max ] as the dominant vibration amplitude, which is expressed by the formula as follows:

[0037]

[0038] A dom (t) represents the dominant vibration amplitude at the t time point, represents a maximum value function, M min represents the lower limit of the frequency band of interest, M max represents the upper limit of the frequency band of interest;

[0039] Obtaining the proportion of each complex spectrum value in the preset frequency band interval [M min ,M max ] in the total complex spectrum value in the preset frequency band interval [M min ,M max ], which is expressed by the formula as follows:

[0040]

[0041] m∈[M min ,M max ];

[0042] P m (t) represents the proportion of the complex spectrum value X(m, t) in the total complex spectrum value in the preset frequency band interval [M min ,M max ], X(i, t) represents a complex spectrum value of the ith frequency band of the complex spectrum vector at the t time point;

[0043] Obtaining the vibration energy entropy based on the proportion, which is expressed by the formula as follows:

[0044]

[0045] H e (t) represents the vibration energy entropy at the t time point.

[0046] Preferably, constructing a material level vibration correlation model based on the material level fluctuation rate and the material level data, which is expressed by the formula as follows:

[0047]

[0048] V pred(t) represents the vibration amplitude of the stone coal bunker at time t, K represents the gain coefficient, β represents the material level fluctuation weight, V0 represents the basic vibration amplitude, γ represents the material level change weight, L f (t-2) represents the filtered material level data at time t-2.

[0049] Preferably, the deviation of the dominant vibration amplitude and the vibration amplitude of the stone coal bunker is calculated, which is expressed by the formula:

[0050]

[0051] In the formula, ε represents the zero protection constant, D v (t) represents the deviation of the dominant vibration amplitude and the vibration amplitude of the stone coal bunker at time t;

[0052] Based on the deviation and the vibration energy entropy, a coupling early warning index is obtained, which is expressed by the formula:

[0053] I c (t) = D v (t) · (1 + η · H e (t));

[0054] In the formula, I c (t) represents the coupling early warning index at time t, and η represents the vibration energy entropy amplification coefficient.

[0055] Preferably, the mean value of the coupling early warning index is calculated, which is expressed by the formula:

[0056]

[0057] In the formula, represents the mean value of the coupling early warning index at time t, and I c (i) represents the coupling early warning index at time i;

[0058] Based on the mean value and the material level fluctuation rate, a warning threshold is obtained, which is expressed by the formula:

[0059]

[0060] In the formula, T c (t) represents the warning threshold at time t, T c0 represents the reference threshold, and λ and μ represent the adjustment coefficients.

[0061] Preferably, based on the coupling early warning index and the warning threshold, a risk level is obtained, which is expressed by the formula:

[0062]

[0063] In the formula, R f (t) represents the risk level at time t.

[0064] The present application has the following advantages:

[0065] 1. The dynamic change of the stone coal bunker level (such as material impact, uneven accumulation) will directly change the internal force balance of the coal mill, causing the change of vibration characteristics. The present application builds a level-vibration correlation model, taking the level fluctuation rate as the "pre-signal" of vibration anomaly, so that the monitoring system can identify the risk before the actual vibration exceeds the standard, avoiding the false alarm caused by the "lag response" of traditional single vibration monitoring.

[0066] 2. The dominant vibration amplitude represents the actual vibration intensity of the equipment, and the stone coal bunker vibration amplitude is calculated by the "reference amplitude" of the correlation model. The present application can quantify the difference between the actual vibration and the expected state by calculating the deviation between the two. A small deviation indicates a stable state, while a large deviation indicates a potential fault. This quantitative mechanism solves the false alarm or missed alarm problem caused by the "one-size-fits-all" of traditional threshold monitoring.

[0067] 3. The vibration energy entropy reflects the degree of disorder of energy distribution in the frequency band. When the equipment has complex multi-frequency vibration (such as bearing wear + structural resonance), the energy entropy increases; single fault (such as single frequency vibration) reduces the energy entropy. The present application can distinguish between "simple fault" and "composite fault" through joint analysis of energy entropy and dominant vibration amplitude, improving the accuracy of risk diagnosis.

[0068] 4. A single parameter (such as vibration amplitude) is easily disturbed, leading to misjudgment. The present application forms a "three-dimensional risk assessment system" by coupling the level fluctuation rate, the deviation of the dominant vibration amplitude, and the vibration energy entropy. Each parameter reflects the equipment state from a different angle: the level fluctuation rate captures the influence of material changes, the deviation quantifies the degree of abnormality, and the energy entropy analyzes the fault type. The fusion of the three makes the risk level classification upgrade from "single threshold judgment" to "multi-dimensional feature matching", realizing the qualitative change from "passive response" to "active early warning". BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 The flowchart of the embodiment method of the present application is shown. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0071] It should be understood that the step numbers used herein are only for the convenience of description, and are not limited to the execution sequence of the steps.

[0072] It is to be understood that the terms used in the specification of the present application are for the purpose of describing particular embodiments and do not intend to limit the present application. As used in the specification and the appended claims of the present application, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0073] The terms "comprise" and "include" indicate the presence of described features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0074] The term "and / or" refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0075] Embodiment one:

[0076] A medium-speed coal mill is usually composed of core components such as grinding plates, grinding rollers, separators, reducers, motors, housings, and slag removal systems. The slag removal system is an important part of the coal mill, responsible for handling the stone coal (coal gangue, impurities, etc. that are not completely ground) generated during the grinding process. The stone coal bin is the terminal equipment of the slag removal system of the medium-speed coal mill, and its main functions include:

[0077] Storage of stone coal: Collecting stone coal generated during the grinding process to avoid environmental pollution and equipment wear caused by direct discharge.

[0078] Transportation and processing: Transporting stone coal to external storage or processing facilities (such as cars and ash yards) through pneumatic transportation (such as negative pressure systems) or mechanical transportation (such as scraper conveyors and bucket elevators).

[0079] Referring to Figure 1 A medium-speed coal mill vibration early warning method based on the stone coal bin level state, comprising the following steps:

[0080] Obtain the level data and vibration data of the medium-speed coal mill, and obtain the level fluctuation rate based on the level data;

[0081] Take the vibration data as the input of the Fast Fourier Transform algorithm FFT, output the complex spectrum vector, and obtain the dominant vibration amplitude and vibration energy entropy based on the complex spectrum vector;

[0082] Construct a level-vibration correlation model based on the level fluctuation rate and level data, and output the vibration amplitude of the stone coal bin;

[0083] Calculate the deviation of the dominant vibration amplitude and the vibration amplitude of the stone coal bin, and obtain the coupling early warning index based on the deviation and the vibration energy entropy;

[0084] calculating a mean value of the coupling early warning index, and obtaining an early warning threshold based on the mean value and the material level fluctuation rate;

[0085] obtaining a risk level based on the coupling early warning index and the early warning threshold, and an operator performing a corresponding protective measure according to the obtained risk level.

[0086] The material level data, the vibration data and the risk level will be finally uploaded to a DCS system to be viewed by an operator, and if the risk level exceeds a preset risk threshold, an alarm will be given, and the material level data, the vibration data and the risk level will be stored, and the operator can further query historical material level data, vibration data and risk level through the DCS system.

[0087] The vibration data includes vibration acceleration, vibration speed and displacement, and is collected by a piezoelectric acceleration sensor installed at a non-driving end bearing seat of a motor or a bearing seat of a speed reducer.

[0088] The material level data includes a material level height, and is obtained by an audio sensor (sonar) installed outside a stone coal bin.

[0089] Preferably, the method further comprises filtering the material level data and the vibration data;

[0090] The filtered material level data is expressed by a formula as follows:

[0091] L f (t)=α L ·L raw (t)+(1-α L )·L f (t-1);

[0092] In the formula, L f (t) represents the filtered material level data at time t, α L represents a first-order low-pass filtering coefficient of the material level data, L raw (t) represents the material level data at time t, and L f (t-1) represents the filtered material level data at time t-1.

[0093] The filtered vibration data is expressed by a formula as follows:

[0094] V f (t)=α V ·V raw (t)+(1-α V )·V f (t-1);

[0095] In the formula, V f (t) represents the filtered vibration data at time t, αV V (t) represents the first-order low-pass filter coefficient of the vibration data, V raw (t) represents the vibration data at time t, V f (t-1) represents the filtered vibration data at time t-1.

[0096] Preferably, the material level fluctuation rate is expressed by the formula:

[0097]

[0098] In the formula, R w (t) represents the material level fluctuation rate at time t, ΔT represents the time window length of the material level data, which is preset by a field expert, N represents the number of sampling points of the time window length, Δt represents the data sampling time interval, and L f (i) represents the filtered material level data at time i, L f (i-1) represents the filtered material level data at time i-1.

[0099] The number of sampling points is expressed by the formula:

[0100]

[0101] Preferably, the vibration data is taken as the input of the fast Fourier transform algorithm FFT, and the specific steps are as follows:

[0102] From time t, N w past time filtered vibration data are taken to construct a vibration window vector, which is expressed by the formula:

[0103]

[0104] In the formula, V window (t) represents the vibration window vector at time t, V f (t-N w +2) represents the filtered vibration data at time t-N w +2.

[0105] The vibration window vector is windowed, which is expressed by the formula:

[0106] x(t) = V window (t)⊙w.

[0107] In the formula, represents the Hadamard product operator, w represents the window function vector, and the Hann window is used in this embodiment, and x(t) represents the vibration window vector at time t after windowing.

[0108] The windowed vibration window vector is taken as the input of the fast Fourier transform algorithm FFT, and a complex spectrum vector is output, which is expressed by the formula:

[0109] X(t) = FFT(x(t));

[0110] In the formula, X(t) represents a complex spectrum vector at t moment, and FFT represents a fast Fourier transform algorithm function;

[0111] The complex spectrum vector includes N w complex spectrum values, and is expressed by a formula as follows:

[0112] X(t) = [X(1,t), …, X(m,t), …, X(N w ,t)] T ;

[0113] In the formula, X(m,t) represents a complex spectrum value of the m frequency band of the complex spectrum vector at t moment;

[0114] The maximum complex spectrum value of the complex spectrum vector in the preset frequency band interval [M min , M max ] is taken as a dominant vibration amplitude, and is expressed by a formula as follows:

[0115]

[0116] In the formula, A dom (t) represents the dominant vibration amplitude at t moment, represents a maximum value function, M min represents a lower limit of a frequency band of interest, which is set by a domain expert according to device characteristics, and M max represents an upper limit of the frequency band of interest, which is set by the domain expert according to the device characteristics;

[0117] The proportion of each complex spectrum value in the preset frequency band interval [M min , M max ] in the total complex spectrum value of the preset frequency band interval [M min , M max ] is obtained, and is expressed by a formula as follows:

[0118]

[0119] m ∈ [M min , M max ];

[0120] In the formula, P m (t) represents the proportion of the complex spectrum value X(m,t) in the total complex spectrum value of the preset frequency band interval [M min , M max ], and X(i,t) represents a complex spectrum value of the i frequency band of the complex spectrum vector at t moment;

[0121] The vibration energy entropy is obtained based on the proportion, and is expressed by a formula as follows:

[0122]

[0123] H e (t) represents the vibration energy entropy at time t.

[0124] Preferably, a material level vibration correlation model is constructed based on the material level fluctuation rate and the material level data, and is expressed by a formula as follows:

[0125]

[0126] V pred (t) represents the vibration amplitude of the stone coal bunker at time t, K represents a gain coefficient, β represents a material level fluctuation weight, V0 represents a basic vibration amplitude, γ represents a material level change weight, L f (t-2) represents the filtered material level data at time t-2;

[0127] represents a vibration component caused by the operation rate. The faster the rate, the larger the value, and the greater the predicted vibration, quantifies a vibration component caused by the operation impact. The more abrupt the operation, the larger the value, and R w (t) represents the instability degree of the material level as a whole. On the basis of a material level that is already violently fluctuating, any new operation is more likely to cause instability and vibration.

[0128] Total vibration = (vibration caused by change rate) + (vibration caused by acceleration) + (basic vibration).

[0129] Preferably, a deviation of the dominant vibration amplitude from the vibration amplitude of the stone coal bunker is calculated, and is expressed by a formula as follows:

[0130]

[0131] ε represents a protection constant against zero, D v (t) represents the deviation of the dominant vibration amplitude from the vibration amplitude of the stone coal bunker at time t;

[0132] A coupling early warning index is obtained based on the deviation and the vibration energy entropy, and is expressed by a formula as follows:

[0133] I c (t) = D v (t) · (1 + η · H e (t));

[0134] I c (t) represents the coupling early warning index at time t, and η represents a vibration energy entropy amplification coefficient.

[0135] Preferably, the average of the coupling early warning index is calculated, which is expressed in the formula as follows:

[0136]

[0137] In the formula, I represents the average of the coupling early warning index at time t, I c (i) represents the coupling early warning index at time i;

[0138] The early warning threshold is obtained based on the average and the material level fluctuation rate, which is expressed in the formula as follows:

[0139]

[0140] In the formula, T c (t) represents the early warning threshold at time t, T c0 represents the reference threshold, and λ and μ represent adjustment coefficients.

[0141] Preferably, the risk level is obtained based on the coupling early warning index and the early warning threshold, which is expressed in the formula as follows:

[0142]

[0143] In the formula, R f (t) represents the risk level at time t.

[0144] The corresponding protective measures include:

[0145] If R f (t) = 1, then:

[0146] The current operating parameters of the coal mill are kept unchanged, and the monitoring and data recording are continued.

[0147] If R f (t) = 2, then:

[0148] The monitoring and early warning prompt are strengthened, the frequency of collecting and recording vibration data is appropriately increased, and it is checked whether the sensor joint is loose and the line is intact.

[0149] If R f (t) = 3, then:

[0150] The operation and maintenance engineer is informed, and it is suggested that the internal inspection of the coal mill be arranged during the nearest planned shutdown period, and the wear conditions of the grinding roller and the liner plate are checked.

[0151] If R f (t) = 4, then:

[0152] The load reduction operation and the preparation for shutdown maintenance are performed, and the operator needs to significantly reduce the load of the coal mill (such as reducing the coal supply amount by more than 20%) and prepare for the shutdown maintenance in the short term.

[0153] If R f (t) = 5, then:

[0154] Immediately stop or emergency stop, notify the site personnel away from the equipment, to prevent possible mechanical damage accident.

[0155] Embodiment two:

[0156] The embodiment provides an electronic device, including a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the vibration early warning method of the medium-speed coal mill based on the stone bunker stock level state according to any embodiment of the present application when executing the program.

[0157] In the embodiments of the present application, "at least one" means one or more, and "multiple" means two or more. "And / or" describes the association relationship of the associated objects, which means that there can be three kinds of relationships, for example, A and / or B, which can represent the cases of A alone, A and B together, and B alone. Wherein A and B can be singular or plural. The character " / " generally represents that the front and rear associated objects are a kind of "or" relationship. "At least one of the following" and the like means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, wherein a, b, c can be single or multiple.

[0158] Those skilled in the art can realize that the units and algorithm steps described in the embodiments disclosed in the present application can be realized by electronic hardware, computer software and combination of electronic hardware and computer software. Whether the functions are realized by hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0159] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0160] In several embodiments provided in the present application, any function, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0161] The above description is only some embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A vibration early warning method for a medium-speed coal mill based on the material level status of a stone coal bunker, characterized in that, Includes the following steps: Obtain material level data and vibration data from a medium-speed coal mill, and obtain the material level fluctuation rate based on the material level data; The vibration data is used as input to the Fast Fourier Transform algorithm, which outputs a complex spectrum vector. The dominant vibration amplitude and vibration energy entropy are obtained based on the complex spectrum vector. Based on the material level volatility and material level data, a material level vibration correlation model is constructed to output the vibration amplitude of the stone coal bunker. Calculate the deviation between the dominant vibration amplitude and the vibration amplitude of the stone coal bunker, and obtain a coupled early warning index based on the deviation and vibration energy entropy; Calculate the mean of the coupled early warning indicators, and obtain the early warning threshold based on the mean and the material level fluctuation rate; The risk level is obtained based on the coupled early warning indicators and early warning thresholds.

2. The vibration early warning method for a medium-speed coal mill based on the material level status of the stone coal bunker according to claim 1, characterized in that, The method further includes filtering the material level data and vibration data; The filtered material level data is expressed by the formula: L f (t)=α L ·L raw (t)+(1-α L )·L f (t-1); In the formula, L f (t) represents the filtered material level data at time t, α L L represents the first-order low-pass filter coefficient for material level data. raw (t) represents the material level data at time t, L f (t-1) represents the filtered material level data at time t-1; The filtered vibration data is expressed by the formula: V f (t)=α V ·V raw (t)+(1-α V )·V f (t-1); In the formula, V f (t) represents the filtered vibration data at time t, α V V represents the first-order low-pass filter coefficient of the vibration data. raw (t) represents the vibration data at time t, V f (t-1) represents the filtered vibration data at time t-1.

3. The vibration early warning method for a medium-speed coal mill based on the material level status of the stone coal bunker according to claim 2, characterized in that, The material level fluctuation rate is expressed by the formula: In the formula, R w (t) represents the level fluctuation rate at time t, ΔT represents the time window length of the level data, N represents the number of sampling points within the time window, Δt represents the data sampling time interval, and L f (i) represents the filtered material level data at time i, L f (i-1) represents the filtered material level data at time i-1.

4. The vibration early warning method for a medium-speed coal mill based on the material level status of the stone coal bunker according to claim 3, characterized in that, The vibration data is used as input to the Fast Fourier Transform (FFT) algorithm. The specific steps are as follows: Starting from time t, take N. w A vibration window vector is constructed from the filtered vibration data of past moments, expressed by the formula: In the formula, V window (t) represents the vibration window vector at time t, V f (tN w +2) indicates tN w Filtered vibration data at time +2; The vibration window vector is windowed, as expressed by the formula: x(t)=V window (t)⊙w; In the formula, ⊙ represents the Hadamard product operator, w represents the window function vector, and x(t) represents the vibration window vector at time t after windowing; The windowed vibration window vector is used as the input to the Fast Fourier Transform (FFT) algorithm, and the output is a complex spectrum vector, expressed by the formula: X(t) = FFT(x(t)); In the formula, X(t) represents the complex spectrum vector at time t, and FFT represents the Fast Fourier Transform algorithm function; The complex spectrum vector includes N w The complex spectrum values ​​are expressed by the formula: X(t)=[X(1,t),…,X(m,t),…,X(N w ,t)] T ; In the formula, X(m,t) represents the complex spectrum value of the m-th frequency band of the complex spectrum vector at time t; Take the complex spectrum vector within the preset frequency band [M] min M max The maximum complex spectral value in the equation is taken as the dominant vibration amplitude, and is expressed by the formula: In the formula, A dom (t) represents the dominant vibration amplitude at time t. M represents the maximum value function. min M represents the lower limit of the frequency band of interest. max This indicates the upper limit of the frequency band of interest; Obtain the preset frequency band [M] min M max The complex spectral values ​​in the [M] range within the preset frequency band [M] min M max The proportion of the total complex spectrum values ​​is expressed by the formula: m∈[M min ,M max ]; In the formula, P m (t) represents the complex spectral value X(m,t) within a preset frequency band [M]. min M max The proportion of the total complex spectrum values, where X(i,t) represents the complex spectrum value of the i-th frequency band of the complex spectrum vector at time t; The vibration energy entropy is obtained based on the aforementioned proportion, expressed by the formula: In the formula, H e (t) represents the vibrational energy entropy at time t.

5. The vibration early warning method for a medium-speed coal mill based on the material level status of the stone coal bunker according to claim 4, characterized in that, Based on the aforementioned material level volatility and material level data, a material level vibration correlation model is constructed, expressed by the following formula: In the formula, V pred (t) represents the vibration amplitude of the stone and coal bunker at time t, K represents the gain coefficient, β represents the weight of material level fluctuation, V0 represents the basic vibration amplitude, γ represents the weight of material level change, and L f (t-2) represents the filtered material level data at time t-2.

6. The vibration early warning method for a medium-speed coal mill based on the material level status of the stone coal bunker according to claim 5, characterized in that, The deviation between the dominant vibration amplitude and the vibration amplitude of the stone and coal bunker is calculated and expressed by the formula: In the formula, ε represents the zero protection constant, and D v (t) represents the deviation between the dominant vibration amplitude and the vibration amplitude of the stone coal bunker at time t; The coupled early warning index is obtained based on the aforementioned deviation and vibration energy entropy, expressed by the formula: I c (t)=D v (t)·(1+η·H e (t)); In the formula, I c (t) represents the coupling early warning index at time t, and η represents the vibration energy entropy amplification coefficient.

7. The vibration early warning method for a medium-speed coal mill based on the material level status of the stone coal bunker according to claim 6, characterized in that, The mean of the coupled early warning indicators is calculated, expressed by the formula: In the formula, I represents the mean of the coupled early warning index at time t. c (i) represents the coupling early warning index at time i; The early warning threshold is obtained based on the mean and the material level fluctuation rate, expressed by the formula: In the formula, T c (t) represents the warning threshold at time t, T c0 λ represents the baseline threshold, and μ represents the adjustment coefficients.

8. The vibration early warning method for a medium-speed coal mill based on the material level status of the stone coal bunker according to claim 7, characterized in that, The risk level is obtained based on the coupled early warning indicators and early warning thresholds, expressed by the formula: In the formula, R f (t) represents the risk level at time t.

Images