Coal mill failure risk assessment method and device

Abstract

The present invention provides a method and device for assessing the risk of coal mill failure, wherein the method comprises: obtaining a first characteristic parameter of a target coal mill; determining the state of the target coal mill based on the first characteristic parameter and a state classification model; when the state of the target coal mill is a faulty state, obtaining a second characteristic parameter of the target coal mill; and assessing the faulty components of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter. The method and device for assessing the risk of coal mill failure provided by the present invention integrate process parameters and equipment parameters to perform coal mill failure risk assessment, and integrate logistic regression and hierarchical analysis to construct a comprehensive coal mill failure diagnosis and location system, which can more accurately determine whether the target coal mill has a fault and more accurately locate the faulty components of the target coal mill.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a method and device for assessing coal mill failure risk. Background Art

[0002] Coal mills are the primary energy supply equipment in cement plants, used to grind raw coal into fine pulverized coal. During the pulverized coal preparation process, the safety and operating conditions of the pulverizer itself impact its safe production and operation. Accurately assessing the failure risks of pulverizers can promptly identify potential safety hazards, reduce operating costs, improve operational stability, and help lower carbon emissions. It also provides a crucial basis for planned maintenance in cement plants, ensuring safe and stable cement production.

[0003] At present, the main fault risk assessment methods for coal mills include: fault diagnosis based on a knowledge base formed by expert experience; using mathematical methods to establish a coal mill quality and energy mechanism model based on real-time monitoring data collected by the DCS (Distributed Control System), and judging faults by calculating the difference between estimated and measured values; using machine learning algorithms to train a coal mill fault diagnosis model based on a large amount of normal and faulty operating condition data, and performing fault diagnosis based on the coal mill fault diagnosis model.

[0004] However, the above methods all have the disadvantages of inaccurate fault type diagnosis, that is, inaccurate fault diagnosis and inaccurate fault component positioning. Summary of the Invention

[0005] The present invention provides a coal mill fault risk assessment method and device, which are used to solve the defect of inaccurate fault diagnosis of coal mill in the prior art and to achieve more accurate assessment of the fault type of the coal mill.

[0006] The present invention provides a coal mill fault risk assessment method, comprising:

[0007] Acquiring a first characteristic parameter of a target coal mill;

[0008] determining the state of the target coal mill based on the first characteristic parameter and a state classification model;

[0009] When the target coal mill is in a fault state, obtaining a second characteristic parameter of the target coal mill;

[0010] evaluating a faulty component of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter;

[0011] Among them, the first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature and inlet pressure of the main motor; the state classification model is a logistic regression model; the second characteristic parameter includes the first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder classifier; the first weight is predetermined based on the hierarchical analysis method.

[0012] According to a coal mill fault risk assessment method provided by the present invention, determining the state of the target coal mill based on the first characteristic parameter and the state classification model includes:

[0013] Inputting the first feature parameter into the state classification model to obtain a first probability;

[0014] When the first probability is greater than a probability threshold, the state of the target coal mill is determined to be a fault state.

[0015] According to a coal mill fault risk assessment method provided by the present invention, the faulty component of the target coal mill is assessed based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter, including:

[0016] obtaining the health of each layer of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter;

[0017] Based on the health status, a faulty component of the target coal mill is evaluated.

[0018] According to a coal mill fault risk assessment method provided by the present invention, obtaining the health of each level of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter includes:

[0019] Obtaining a second weight corresponding to the health level of the current level based on the health level of the current level and the first weight corresponding to the health level of the current level;

[0020] The health of the previous level is obtained based on the health of the current level and the second weight corresponding to the health of the current level.

[0021] According to a coal mill fault risk assessment method provided by the present invention, obtaining a second weight corresponding to the health degree of the current level based on the health degree of the current level and the first weight corresponding to the health degree of the current level includes:

[0022] Based on each group of health degrees at the current level, a first weight corresponding to each group of health degrees, and a preset balancing function, a second weight corresponding to each group of health degrees is obtained.

[0023] According to a coal mill failure risk assessment method provided by the present invention, the balance function is an exponential function.

[0024] The present invention also provides a coal mill fault risk assessment device, comprising:

[0025] A first acquisition module, configured to acquire a first characteristic parameter of a target coal mill;

[0026] a classification module, configured to determine the state of the target coal mill based on the first characteristic parameter and a state classification model;

[0027] a second acquisition module, configured to acquire a second characteristic parameter of the target coal mill when the target coal mill is in a fault state;

[0028] an evaluation module, configured to evaluate a faulty component of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter;

[0029] Among them, the first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature and inlet pressure of the main motor; the state classification model is a logistic regression model; the second characteristic parameter includes the first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder classifier; the first weight is predetermined based on the hierarchical analysis method.

[0030] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any one of the coal mill fault risk assessment methods described above is implemented.

[0031] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the coal mill failure risk assessment method as described above is implemented.

[0032] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned coal mill failure risk assessment methods.

[0033] The coal mill fault risk assessment method and device provided by the present invention utilizes a logistic regression model to preliminarily classify the target coal mill's status based on its first characteristic parameter. If the target coal mill is in a faulty state, the device locates faulty components at different levels of the target coal mill based on its second characteristic parameter. This method integrates process parameters and equipment parameters to perform coal mill fault risk assessment, integrating logistic regression with the analytic hierarchy process to construct a comprehensive coal mill fault diagnosis and location system. This system can more accurately determine whether a target coal mill is faulty and locate faulty components in the target coal mill. Furthermore, the system calculates the health of key mechanisms such as the transmission system, grinding system, and separation system using electrical and vibration data, enabling real-time dynamic interpretation of the results of the process parameter-based logistic regression fault diagnosis algorithm. This improves the interpretability of the coal mill fault diagnosis algorithm and enhances the localizability of coal mill fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 This is one of the flow charts of the coal mill failure risk assessment method provided by the present invention;

[0036] Figure 2 Schematic diagram of the coal mill failure risk assessment index system in the coal mill failure risk assessment method provided by the present invention;

[0037] Figure 3 This is the second flow chart of the coal mill failure risk assessment method provided by the present invention;

[0038] Figure 4 It is a structural schematic diagram of the coal mill fault risk assessment device provided by the present invention;

[0039] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0040] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0041] In the description of the embodiments of the present invention, the terms "first", "second" and "third" are used for descriptive purposes only and should not be understood as indicating or implying relative importance, and are not related to the order.

[0042] The following combination Figures 1 to 5 The present invention describes a coal mill fault risk assessment method and device.

[0043] Figure 1 This is one of the flow charts of the coal mill failure risk assessment method provided by the present invention. Figure 1 As shown, the coal mill failure risk assessment method provided by the embodiment of the present invention may be executed by a coal mill failure risk assessment device. The method includes: step 101, step 102, step 103 and step 104.

[0044] Step 101: Obtain a first characteristic parameter of a target coal mill.

[0045] The first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature and inlet pressure of the main motor.

[0046] Specifically, the target coal mill is a coal mill that needs to be evaluated for failure risk.

[0047] According to the changing characteristics of each process parameter when the coal mill fails, the characteristic parameters of a representative coal mill can be used as the first characteristic parameters of the target coal mill.

[0048] Optionally, at least one of the current, outlet pressure, outlet temperature and inlet pressure of the main motor of the target coal mill may be obtained as the first characteristic parameter of the target coal mill.

[0049] Step 102: Determine the state of the target coal mill based on the first characteristic parameter and the state classification model.

[0050] Among them, the state classification model is a logistic regression model.

[0051] Specifically, the logistic regression model is a model derived from the logistic regression method. The state classification model is used to classify the state of the target coal mill based on the first characteristic parameter. The logistic regression method is a supervised machine learning algorithm that is widely used in fault classification.

[0052] The state of the target coal mill may include a normal state and a fault state.

[0053] By inputting the first characteristic parameter of the target coal mill into the state classification model, it can be determined whether the state of the target coal mill is a normal state or a fault state.

[0054] It is understood that a logistic regression analysis can be performed on the sample data based on a logistic regression method to obtain a state classification model. The sample data may include the first characteristic parameter and state of the sample coal mill. There may be multiple sample coal mills. Coal mills at different times can be considered different sample coal mills.

[0055] Step 103: When the target coal mill is in a fault state, obtain a second characteristic parameter of the target coal mill.

[0056] The second characteristic parameters include the first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder classifier.

[0057] Specifically, when the state of the target coal mill is a fault state, the faulty components of the target coal mill may be further evaluated.

[0058] The first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder classifier can be obtained as the second characteristic parameters of the target coal mill.

[0059] A coal mill consists of a transmission system, a grinding system, and a separation system. The transmission system primarily consists of a main motor and a main reducer. The grinding system primarily consists of grinding rollers and a pressurizing device. The separation system primarily consists of a classifier.

[0060] The first parameter of the main motor can be used to characterize the health of the main motor. The second parameter of the main reducer can be used to characterize the health of the main reducer. The health of the main motor and the health of the main reducer can be used to characterize the health of the transmission system.

[0061] The third parameter of the grinding roller can be used to characterize the health of the grinding roller. The fourth parameter of the pressure device can be used to characterize the health of the pressure device. The health of the grinding roller and the health of the pressure device can be used to characterize the health of the grinding system.

[0062] The fifth parameter of the powder classifier can be used to characterize the health of the powder classifier. The health of the powder classifier can be used to characterize the health of the separation system.

[0063] The health of the transmission system, the health of the grinding system and the health of the separation system can be used to characterize the health of the coal mill.

[0064] Optionally, the first parameter of the main motor may include at least one of the current, first vibration parameter, sound pressure level, bearing temperature, stator temperature, etc. The first vibration parameter may include at least one of amplitude and vibration frequency, etc.

[0065] Optionally, the second parameter of the main reducer may include at least one of the thrust shoe oil tank temperature, the second vibration parameter, and the gearbox oil pool temperature. The second vibration parameter may include at least one of the amplitude and the vibration frequency.

[0066] Optionally, the third parameter of the grinding roller may include the sealing air pressure outlet pressure.

[0067] Optionally, the fourth parameter of the pressurizing device may include at least one of the hydraulic oil station tank temperature and the grinding roller loading pressure.

[0068] Optionally, the fifth parameter of the powder classifier may include at least one of a third vibration parameter of the powder classifier motor, a current of the powder classifier motor, and a fourth vibration parameter of the powder classifier reducer. The third vibration parameter may include at least one of amplitude and frequency. The fourth vibration parameter may include at least one of amplitude and frequency.

[0069] It is understandable that after step 102, when the state of the target coal mill is normal, the process may return to step 101 and re-evaluate the coal mill failure risk to implement monitoring of the coal mill failure risk.

[0070] Step 104: Evaluate the faulty component of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter.

[0071] The first weight is predetermined based on the analytic hierarchy process.

[0072] Specifically, based on the second characteristic parameter and the AHP method, a health analysis of the target coal mill can be performed layer by layer, a comprehensive assessment of the target coal mill's failure risk can be performed, and the specific component of the target coal mill currently experiencing a failure can be located. This component is referred to as the faulty component. The faulty component can include components at different levels.

[0073] For example, when it is determined that there is a fault in the main reducer, the faulty components at different levels of the target coal mill can be the transmission system and the main reducer respectively; when it is determined that there is a fault in the grinding roller, the faulty components at different levels of the target coal mill can be the grinding system and the grinding roller respectively.

[0074] It is understandable that the first weight of the health level at each level can be determined in advance based on the hierarchical analysis method. The first weight of the health level at each level includes the first weight corresponding to each second characteristic parameter.

[0075] Optionally, the first weight may be calculated based on the analytic hierarchy process and entropy weight.

[0076] It's important to note that traditional expert system approaches rely too heavily on knowledge bases and are unable to identify problems outside of them. Mechanism-based models are unable to address model mismatches caused by the variability of coal types in actual cement plant operations. Machine learning-based approaches require large amounts of hard-to-obtain fault data to support model training and are unable to explain the causes of faults.

[0077] The coal mill failure risk assessment method provided by the embodiment of the present invention is as follows:

[0078] The embodiment of the present invention uses a logistic regression model to preliminarily classify the state of a target coal mill based on its first characteristic parameter. If the target coal mill is in a faulty state, the target coal mill's faulty components at different levels are located based on its second characteristic parameter. The coal mill fault risk assessment is performed by integrating process parameters and equipment parameters. A comprehensive coal mill fault diagnosis and location system is constructed by integrating logistic regression with the analytic hierarchy process. This system can more accurately determine whether a target coal mill has a fault and more accurately locate its faulty components. Furthermore, the health of key mechanisms such as the transmission system, grinding system, and separation system is calculated using electrical and vibration data. This allows for real-time dynamic interpretation of the results of the process parameter-based logistic regression fault diagnosis algorithm, improving the interpretability of the coal mill fault diagnosis algorithm and enhancing the localizability of coal mill fault diagnosis.

[0079] Based on the content of any of the above embodiments, determining the state of the target coal mill based on the first characteristic parameter and the state classification model includes: inputting the first characteristic parameter into the state classification model to obtain a first probability.

[0080] Specifically, the first characteristic parameter of the target coal mill can be input into the state classification model to obtain a first probability. The first probability is used to indicate whether the target coal mill has a coal shortage fault.

[0081] Optionally, the first characteristic parameters of the target coal mill may be standardized and then input into the state classification model to eliminate the adverse effects of large numerical differences among the first characteristic parameters on the classification accuracy of the state classification model, thereby improving the accuracy of state classification.

[0082] Optionally, for each first characteristic parameter, the following formula may be used for normalization:

[0083]

[0084] Where X represents the first characteristic parameter before normalization; represents the mean of X, represents the standard deviation of X; X * is the first characteristic parameter after standardization. and It can be obtained based on the first sample data.

[0085] Optionally, when the first characteristic parameter includes four parameters, namely, the current, outlet pressure, outlet temperature, and inlet pressure of the main motor, the first characteristic parameter may be expressed as the following characteristic vector X(t):

[0086] X(t)=[x1(t),x2(t),x3(t),x4(t)] (2)

[0087] Among them, x1(t), x2(t), x3(t), and x4(t) represent the current, outlet pressure, outlet temperature, and inlet pressure of the main motor at time t, respectively.

[0088] Optionally, the model function of the state classification model may be, but is not limited to, a Sigmoid function. The calculation formula of the Sigmoid function is as follows:

[0089]

[0090] Optionally, when the first characteristic parameter includes four parameters, namely, the current, outlet pressure, outlet temperature, and inlet pressure of the main motor, the calculation formula of the first probability can be:

[0091]

[0092] Among them, y t =1 and y t =0 represents coal outage and non-coal outage respectively; α represents the regression intercept; β represents the regression coefficient; x1, x2, x3, x4 represent the current, outlet pressure, outlet temperature and inlet pressure of the main motor respectively.

[0093] Optionally, the regression intercept α and the regression coefficient β may be obtained based on the first sample data by using an algorithm such as maximum likelihood estimation.

[0094] We can first perform logarithmic transformation on formula (4):

[0095]

[0096] Let α = β = δ, and substitute into formula (5) to obtain the log-likelihood equation, which is calculated as follows:

[0097] lnL(δ)=∑[yδX(t)-ln(1+exp(δX(t)))] (6)

[0098] Optionally, the regression intercept α and the regression coefficient β may be calculated using a gradient descent method.

[0099] After obtaining the regression intercept α and regression coefficient β, the state classification model can be obtained.

[0100] When the first probability is greater than the probability threshold, the state of the target coal mill is determined to be a fault state.

[0101] Specifically, after obtaining the first probability, the first probability may be compared with a predetermined probability threshold.

[0102] When the first probability is greater than the probability threshold, the state of the target coal mill may be determined to be a fault state.

[0103] When the first probability is less than or equal to the probability threshold, the state of the target coal mill may be determined to be a normal state.

[0104] The probability threshold can be predetermined based on actual conditions. The embodiment of the present invention does not specifically limit the specific value of the probability threshold. For example, the probability threshold can be 0.8 or 0.9, etc.

[0105] The embodiment of the present invention obtains a first probability by inputting a first characteristic parameter into a state classification model. When the first probability is greater than a probability threshold, the state of the target coal mill is determined to be a fault state, thereby more accurately determining whether the target coal mill has a fault.

[0106] Based on the content of any of the above embodiments, based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter, the faulty components of the target coal mill are evaluated, including: based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter, the health of each level of the target coal mill is obtained.

[0107] Specifically, the principle of coal mill fault risk location assessment is to be able to truly reflect the health status of the coal mill. Combined with the analysis of coal mill fault characteristics and status monitoring parameters, parameters that are easy to measure and can better reflect the coal mill status are selected, and a coal mill fault risk assessment index system is established from the system level, component level, and parameter level. This index system analyzes the coal mill fault location based on the three main mechanisms of the coal mill transmission system, grinding system, and separation system, and then constructs a quantifiable and real-time coal mill fault risk assessment index system, such as Figure 2 shown.

[0108] Based on the second characteristic parameter and the Analytic Hierarchy Process (AHP), the target coal mill's health can be analyzed layer by layer to determine its health at each level. The health levels, from top to bottom, are: whole machine level, system level, component level, and parameter level. The system level health includes the health of the transmission system, the grinding system, and the separation system. The component level health includes the health of the main motor, main reducer, grinding roller, pressurizing device, and classifier. The parameter level health can be determined based on the target coal mill's second characteristic parameter.

[0109] It can be understood that the first parameter of the main motor, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressure device, and the fifth parameter of the powder classifier are 5 groups of health at the parameter layer; the health of the main motor, the health of the main reducer, the health of the grinding roller, the health of the pressure device, and the health of the powder classifier are 3 groups of health at the component layer; the health of the transmission system, the health of the grinding system, and the health of the separation system constitute 1 group of health at the system layer.

[0110] Optionally, based on the health of the current level and the first weight corresponding to the health of the current level, a weighted sum of the health of the current level may be obtained as the health of the previous level.

[0111] Optionally, based on each health metric group at the current level and the first weight corresponding to each health metric group, a weighted sum of the health metrics of each group can be obtained as the health metric of the previous level represented by this health metric group. For example, the health of a target coal mill is the health metric of the previous level represented by the health of the transmission system, the health of the grinding system, and the health of the separation system; the health of the grinding system is the health metric of the previous level represented by the health of the grinding rollers and the health of the pressurizing device; and the health of a classifier is the health metric of the previous level represented by the fifth parameter of the classifier.

[0112] Based on the health status, the faulty components of the target coal mill are evaluated.

[0113] Specifically, a comprehensive assessment can be performed based on the health of each level of the target coal mill, and the specific component of the target coal mill currently experiencing a fault can be located. This component is the faulty component. The faulty component can include components at different levels.

[0114] This embodiment of the present invention locates faulty components at different levels of the target coal mill based on the second characteristic parameter of the target coal mill, enabling more accurate localization of faulty components. Furthermore, by calculating the health of key mechanisms such as the transmission system, grinding system, and separation system using electrical and vibration data, the results of a logistic regression fault diagnosis algorithm based on process parameters can be dynamically interpreted in real time. This improves the interpretability of the coal mill fault diagnosis algorithm and enhances the localizability of coal mill fault diagnosis.

[0115] Based on the content of any of the above embodiments, based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter, the health of each level of the target coal mill is obtained, including: based on the health of the current level and the first weight corresponding to the health of the current level, obtaining the second weight corresponding to the health of the current level.

[0116] Optionally, based on the variable weight theory, the first weight corresponding to the health level of the current level may be modified to obtain a second weight corresponding to the health level of the current level.

[0117] Specifically, based on the variable weight theory, according to each group of health degrees at this level, the first weight corresponding to the group of health degrees may be modified to obtain the second weight corresponding to the group of health degrees.

[0118] The health of the previous level is obtained based on the health of the current level and the second weight corresponding to the health of the current level.

[0119] Specifically, based on the health of the current level and the second weight corresponding to the health of the current level, a weighted sum of the health of the current level may be obtained as the health of the previous level.

[0120] Optionally, based on each group of health metrics at the current level and the second weight corresponding to the group of health metrics, a weighted sum of the group of health metrics may be obtained as the health metrics of the previous level represented by the group of health metrics.

[0121] The variable weight theory can highlight the degree of deviation of each parameter and reflect the balance of the parameter status. Use the variable weight theory to correct the first weight, so that the weight corresponding to the parameter changes in time with its status, and make up for the "short board effect" of abnormal parameters. Although the first weight can already reflect the relative importance of each parameter, it still cannot track the fluctuation of the parameter in time. For example, when the value of a parameter of a key component of a coal mill deviates seriously from the standard value, it often means that there is a big problem with the health status of the component. If a constant weight (which can refer to the first weight) is used for evaluation, when other indicators do not change or change slightly, it may mask the negative effect of the indicator, causing the evaluation result to show a normal state, and the ability to track the state of the coal mill will be lost.

[0122] In an embodiment of the present invention, a second weight corresponding to the health of the current level is obtained based on the health of the current level and the first weight corresponding to the health of the current level. The health of the previous level is obtained based on the health of the current level and the second weight corresponding to the health of the current level. The weight corresponding to the health is corrected using the variable weight theory, thereby more accurately locating the faulty component of the target coal mill.

[0123] Based on the content of any of the above embodiments, based on the health of this level and the first weight corresponding to the health of this level, the second weight corresponding to the health of this level is obtained, including: based on each group of health of this level, the first weight corresponding to each group of health and a preset balance function, the second weight corresponding to each group of health is obtained.

[0124] Specifically, based on each group of health levels at this level, the first weight corresponding to each group of health levels, and a preset balancing function, a formula for obtaining the second weight corresponding to each group of health levels may be as follows:

[0125]

[0126] Where s(x) represents the equilibrium function; w″ i Indicates health x i The corresponding first weight; w i Indicates health x i The corresponding second weight; x j Indicates health x i Any health degree included in the set of health degrees to which it belongs; n represents the health degree x i The number of health states included in the health state group.

[0127] The balancing function can be used to amplify the degree to which a parameter deviates from a standard value. The balancing function can be predetermined based on actual conditions. The specific form of the balancing function is not specifically limited in this embodiment of the present invention. Optionally, the balancing function can be an exponential function or a power function.

[0128] The embodiment of the present invention obtains a second weight corresponding to each group of health degrees based on each group of health degrees at this level, the first weight corresponding to each group of health degrees, and a preset balancing function. The second weight can indicate the degree to which the health degrees deviate from the standard value, thereby more accurately locating the faulty components of the target coal mill.

[0129] Based on the content of any of the above embodiments, the balancing function is an exponential function.

[0130] Specifically, the balancing function can be an exponential function, thereby establishing a "penalty-type" state variable weight vector based on the exponential function. For indicators that deviate from the normal value, the weight is adjusted according to the degree of deviation of the indicator beyond the limit. The more serious the indicator deviation, the larger the corresponding variable weight (i.e., the second weight), and the corresponding variable weight is also larger.

[0131] Alternatively, the formula of the equalization function can be as follows:

[0132]

[0133] Among them, e represents the natural constant; x0 represents the health degree x i The penalty threshold (i.e. health x i The standard value of x max 、x min Health x i The upper and lower limits of the threshold interval; β represents the penalty level (β>0). The larger β is, the more obvious the penalty effect is, that is, even a small change in health will cause fluctuations in the evaluation results.

[0134] The embodiment of the present invention uses an exponential function to correct the first weight and obtain a second weight. The second weight can indicate the degree to which the health level deviates from the standard value, thereby more accurately locating the faulty component of the target coal mill.

[0135] To facilitate understanding of the above embodiments of the present invention, the implementation process of the coal mill fault risk assessment method is described below. Figure 3 This is the second flow chart of the coal mill failure risk assessment method provided by the present invention. Figure 3 The implementation process of the coal mill failure risk assessment method is shown.

[0136] The fault characteristics of the coal mill process parameters can be analyzed to select an appropriate first characteristic parameter. After selecting the appropriate first characteristic parameter, a coal mill logistic regression state classification model (i.e., the aforementioned state classification model) can be established based on the first characteristic parameter. Based on this state classification model, the coal mill state can be determined.

[0137] It is also possible to analyze the fault characteristics of the coal mill equipment parameters and select appropriate second characteristic parameters. After selecting the appropriate second characteristic parameters, a coal mill fault risk assessment index system can be established based on the second characteristic parameters.

[0138] Based on the coal mill risk assessment index system established above, a risk assessment index weight determination method combining expert experience, entropy weight and variable weight can be used to solve the problem of selecting index weight coefficients, and combined with the health calculation model, a comprehensive assessment of the coal mill fault risk location in actual operation can be carried out.

[0139] After establishing the coal mill failure risk assessment index system, the judgment matrix can be constructed in combination with expert opinions.

[0140] Based on the process experience of coal mills in cement plants, various indicators at the whole machine level, system level, component level and parameter level can be compared layer by layer to obtain the judgment matrix P of the hierarchical analysis method.

[0141] Each set of health can have a judgment matrix P. The judgment matrix P can be calculated as follows:

[0142]

[0143] Where n represents the number of health levels in each health level group. The value of each element in the judgment matrix P can be determined based on the expert's score.

[0144] After constructing the judgment matrix, the eigenvector of the judgment matrix can be obtained. After obtaining the eigenvector of the judgment matrix, the consistency ratio can be calculated.

[0145] The eigenvector W'=(W'1, W'2, ..., W') of the judgment matrix P can be solved by the square root method. n ), calculated according to the following formula, and the consistency was tested by calculating the consistency ratio.

[0146]

[0147] Among them, w′ i Indicates the eigenvalue of the judgment matrix P, the health degree x in the group of health degrees i A corresponding eigenvalue.

[0148] When the consistency ratio is less than the threshold (which may not be limited to Figure 3 In the case of 0.1) shown in the figure, the health degree x i The corresponding eigenvalues ​​are used as AHP (Analytic Hierarchy Process) weights, thereby obtaining AHP weights.

[0149] When the consistency ratio is greater than or equal to the threshold, the expert opinions are recombined to construct the judgment matrix.

[0150] After establishing the coal mill failure risk assessment index system, the index entropy weight calculation can be performed.

[0151] Information entropy ("entropy") can be used to measure the degree of disorder in data. The smaller the information entropy of a parameter, the greater its dispersion, the greater its variability, and the greater the amount of information it reveals. Consequently, its importance is also greater, and its weight should be greater; conversely, its weight should be smaller.

[0152] If a certain health degree has n data, the information entropy of the health degree can be calculated as follows:

[0153]

[0154]

[0155] Among them, x i and x j Each represents one of the n data above; P i Indicates the normalized value of the data corresponding to the health level.

[0156] The entropy weight of the health It can be calculated as follows:

[0157]

[0158] Here, k represents the number of health degrees included in the set of health degrees to which the health degree belongs.

[0159] After obtaining the AHP weight and the entropy weight, the AHP-entropy weight combination weight can be obtained based on the AHP weight and the entropy weight.

[0160] By combining AHP weights and entropy weights, the formula for forming combined weights can be shown as follows:

[0161]

[0162] Among them, w i " indicates health x i The combined weight of i The corresponding first weight; w i 'Indicates health x i The AHP weight is the weight obtained by the analytic hierarchy process; Indicates health x i The entropy weight of .

[0163] The calculation formula of parameter α is as follows:

[0164]

[0165] Where n represents the health level x i The number of health degrees included in the health degree group; w1,w2,w3,…,w n Represents the health degree x obtained by the analytic hierarchy process i The AHP weights of the health degrees of a group are rearranged from small to large.

[0166] After establishing the coal mill failure risk assessment index system, a suitable function can be selected as the equilibrium function s(x).

[0167] Based on the balance function s(x) and the AHP-entropy weight combination weight, the variable weight comprehensive weight, that is, the second weight, can be obtained.

[0168] After establishing the coal mill failure risk assessment indicator system, the health of the parameter layer can be obtained. The health of the parameter layer can be obtained based on the second characteristic parameter.

[0169] In any of the above embodiments of the present invention, health is used as an indicator for evaluating device status. This indicator refers to the ability of a device to continuously perform its intended function within specified time and conditions. Generally, the closer the device parameters are to their standard values, the better the performance and the higher the health.

[0170] Optionally, for any second characteristic parameter, the healthiness of the second characteristic parameter may be calculated using the following formula:

[0171]

[0172] Where x, x0 represent the measured value and standard value of the second characteristic parameter respectively; X represents the health of the second characteristic parameter; x max 、x min Respectively represent the upper and lower limits (ie, the maximum value and the minimum value) of the threshold of the second characteristic parameter.

[0173] Figure 2 The standard value, maximum value, and minimum value of each second characteristic parameter shown may be as shown in Table 1.

[0174] Table 1 Standard value, maximum value and minimum value of each second characteristic parameter

[0175]

[0176] Among them, the second characteristic parameters A11, A12, A13, A14, A15, and A16 respectively represent the current of the main motor, the vibration amplitude of the main motor, the sound pressure level of the main motor, the front bearing temperature of the main motor, the rear bearing temperature of the main motor, and the stator temperature of the main motor; the second characteristic parameters A21, A22, and A23 respectively represent the thrust bearing oil tank temperature of the main reducer, the vibration amplitude of the main reducer, and the gearbox oil pool temperature of the main reducer; the second characteristic parameter B11 represents the sealing air pressure outlet pressure of the grinding roller; the second characteristic parameters B21 and B22 respectively represent the hydraulic oil station box temperature of the pressurizing device and the grinding roller loading pressure of the pressurizing device; the second characteristic parameters C11, C12, and C13 respectively represent the vibration amplitude of the classifier motor, the current of the classifier motor, and the vibration amplitude of the classifier reducer.

[0177] After obtaining the health of the parameter layer, the health calculation considering the comprehensive weight can be performed layer by layer based on the health of the parameter layer and the comprehensive weight of each variable weight. The status of the main components of the coal mill can be comprehensively analyzed from various levels and multiple aspects to obtain the health of the component layer, system layer and whole machine layer.

[0178] The calculation formulas for component, system, and machine health are as follows:

[0179]

[0180] Among them, x A11 、x A12 、x A13 、x A14 、x A15 、x A16 Respectively represent the health of the second characteristic parameters A11, A12, A13, A14, A15, and A16, which are all the health of the parameter layer; A11 、w A12 、w A13 、w A14 、w A15 、w A16 Represents x A11 、x A12 、x A13 、x A14 、x A15 、x A16 The corresponding second weight; X A1Indicates the health of the main motor, which is the health of the component layer, and is x A11 、x A12 、x A13 、x A14 、x A15 、x A16 The health of the previous level; x A21 、x A22 、x A23 Respectively represent the health of the second characteristic parameters A21, A22, and A23, which are all the health of the parameter layer; A21 、w A22 、w A23 Represents x A21 、x A22 、x A23 The corresponding second weight; X A2 Indicates the health of the main reducer, which is the health of the component layer, and is x A21 、x A22 、x A23 The health of the previous level; x B11 Indicates the health of the second characteristic parameter B11, which is the health of the parameter layer; X B1 Indicates the health of the grinding roller, which is the health of the component layer, and is x B11 The health of the previous level; x B21 、x B22 Respectively represent the health of the second characteristic parameters B21 and B22, both of which are the health of the parameter layer; B21 、w B22 Represents x B21 、x B22 The corresponding second weight; X B2 Indicates the health of the pressurizing device, which is the health of the component layer, and is x B21 、x B22 The health of the previous level; x C11 、x C12 、x C13 Respectively represent the health of the second characteristic parameters C11, C12, and C13, which are all the health of the parameter layer; w C11 、w C12 、w C13 Represents x C11 、x C12 、x C13 The corresponding second weight; X C1 Indicates the health of the powder separator, which is the health of the component layer, and is x C11 、x C12 、x C13 The health level of the previous level; w A1 、w A2 Represents XA1 and X A2 The corresponding second weight; X A Indicates the health of the transmission system, which is the health of the system layer, and is X A1 and X A2 The health level of the previous level; w B1 and w B2 Represents X B1 and X B2 The corresponding second weight; X B Indicates the health of the grinding system, which is the health of the system layer, and is X B1 and X B2 The health of the previous level; X C Indicates the health of the separation system, which is the health of the system layer, and is X C1 The health level of the previous level; w A 、w B 、w C Represents X A 、X B and X C The corresponding second weight; X coal milll Indicates the health of the target coal mill, which is the health of the entire machine layer, and is X A 、X B and X C The health of the next level.

[0181] Based on the status of the coal mill and the health of the component level, system level, and entire machine level, the coal mill failure risk can be located and assessed.

[0182] The coal mill failure risk assessment device provided by the present invention is described below. The coal mill failure risk assessment device described below and the coal mill failure risk assessment method described above can be referenced to each other.

[0183] Figure 4 This is a schematic diagram of the structure of the coal mill fault risk assessment device provided by the present invention. Based on the content of any of the above embodiments, such as Figure 4 As shown, the apparatus includes a first acquisition module 401, a classification module 402, a second acquisition module 403 and an evaluation module 404, wherein:

[0184] A first acquisition module 401 is used to acquire a first characteristic parameter of a target coal mill;

[0185] A classification module 402 is configured to determine the state of the target coal mill based on the first characteristic parameter and the state classification model;

[0186] The second acquisition module 403 is used to acquire a second characteristic parameter of the target coal mill when the target coal mill is in a fault state;

[0187] An evaluation module 404 is configured to evaluate a faulty component of a target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter;

[0188] Among them, the first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature and inlet pressure of the main motor; the state classification model is a logistic regression model; the second characteristic parameter includes the first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder classifier; the first weight is predetermined based on the hierarchical analysis method.

[0189] Specifically, the first acquisition module 401 , the classification module 402 , the second acquisition module 403 and the evaluation module 404 may be electrically connected in sequence.

[0190] The first acquisition module 401 acquires at least one of the current, outlet pressure, outlet temperature, and inlet pressure of a main motor of a target coal mill as a first characteristic parameter of the target coal mill.

[0191] The classification module 402 inputs the first characteristic parameter of the target coal mill into the state classification model to determine whether the state of the target coal mill is a normal state or a fault state.

[0192] The second acquisition module 403 can acquire the first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder concentrator as the second characteristic parameters of the target coal mill.

[0193] The evaluation module 404 can perform a hierarchical health analysis of the target coal mill based on the second characteristic parameter and the analytic hierarchy process, comprehensively assess the failure risk of the target coal mill, and locate the specific component of the target coal mill that is currently experiencing a failure. This component is referred to as the faulty component. The faulty component can include components at different levels.

[0194] Optionally, the classification module 402 may be specifically configured to input the first characteristic parameter into a state classification model to obtain a first probability; and when the first probability is greater than a probability threshold, determine the state of the target coal mill as a fault state.

[0195] Optionally, the evaluation module 404 may include:

[0196] An acquiring unit, configured to acquire the health of each layer of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter;

[0197] The evaluation unit is used to evaluate the faulty components of the target coal mill based on the health status.

[0198] Optionally, the acquisition unit may include:

[0199] a first acquiring subunit, configured to acquire a second weight corresponding to the health level of the current level based on the health level of the current level and the first weight corresponding to the health level of the current level;

[0200] The second acquiring subunit is configured to acquire the health of a previous level based on the health of the current level and a second weight corresponding to the health of the current level.

[0201] Optionally, the first acquisition subunit may be specifically configured to acquire a second weight corresponding to each group of health levels based on each group of health levels at the current level, a first weight corresponding to each group of health levels, and a preset balancing function.

[0202] Optionally, the balance function is an exponential function.

[0203] The coal mill failure risk assessment device provided in an embodiment of the present invention is used to execute the above-mentioned coal mill failure risk assessment method of the present invention. Its implementation method is consistent with the implementation method of the coal mill failure risk assessment method provided by the present invention and can achieve the same beneficial effects, which will not be repeated here.

[0204] The coal mill failure risk assessment device is used in the coal mill failure risk assessment method in the aforementioned embodiments. Therefore, the description and definition of the coal mill failure risk assessment method in the aforementioned embodiments can be used to understand the various execution modules in the embodiments of the present invention.

[0205] The embodiment of the present invention uses a logistic regression model to preliminarily classify the state of a target coal mill based on its first characteristic parameter. If the target coal mill is in a faulty state, the target coal mill's faulty components at different levels are located based on its second characteristic parameter. The coal mill fault risk assessment is performed by integrating process parameters and equipment parameters. A comprehensive coal mill fault diagnosis and location system is constructed by integrating logistic regression with the analytic hierarchy process. This system can more accurately determine whether a target coal mill has a fault and more accurately locate its faulty components. Furthermore, the health of key mechanisms such as the transmission system, grinding system, and separation system is calculated using electrical and vibration data. This allows for real-time dynamic interpretation of the results of the process parameter-based logistic regression fault diagnosis algorithm, improving the interpretability of the coal mill fault diagnosis algorithm and enhancing the localizability of coal mill fault diagnosis.

[0206] Figure 5 Schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5As shown, the electronic device may include: a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 communicate with each other via the communications bus 540. The processor 510 may call logic instructions in the memory 530 to execute a coal mill fault risk assessment method, which includes: obtaining a first characteristic parameter of a target coal mill; determining the state of the target coal mill based on the first characteristic parameter and a state classification model; if the state of the target coal mill is a faulty state, obtaining a second characteristic parameter of the target coal mill; and assessing the faulty component of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter. The first characteristic parameter includes at least one of the current of the main motor, the outlet pressure, the outlet temperature, and the inlet pressure; the state classification model is a logistic regression model; the second characteristic parameter includes a first parameter of the main motor of the target coal mill, a second parameter of the main reducer, a third parameter of the grinding roller, a fourth parameter of the pressurizing device, and a fifth parameter of the classifier; and the first weight is predetermined based on the analytic hierarchy process.

[0207] In addition, the logic instructions in the above-mentioned memory 530 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0208] The processor 510 in the electronic device provided in the embodiment of the present application can call the logic instructions in the memory 530. Its implementation method is consistent with the implementation method of the coal mill fault risk assessment method provided in the present application, and can achieve the same beneficial effects, which will not be repeated here.

[0209] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the coal mill failure risk assessment method provided by the above methods, the method including: obtaining a first characteristic parameter of the target coal mill; determining the state of the target coal mill based on the first characteristic parameter and a state classification model; when the state of the target coal mill is a fault state, obtaining a second characteristic parameter of the target coal mill; evaluating the faulty component of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter; wherein the first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature and inlet pressure of the main motor; the state classification model is a logistic regression model; the second characteristic parameter includes the first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder classifier; the first weight is predetermined based on the hierarchical analysis method.

[0210] When the computer program product provided in the embodiment of the present application is executed, the above-mentioned coal mill failure risk assessment method is implemented. Its specific implementation method is consistent with the implementation method described in the embodiment of the aforementioned method and can achieve the same beneficial effects, which will not be repeated here.

[0211] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the above-mentioned coal mill fault risk assessment method, the method comprising: obtaining a first characteristic parameter of a target coal mill; determining the state of the target coal mill based on the first characteristic parameter and a state classification model; when the state of the target coal mill is a fault state, obtaining a second characteristic parameter of the target coal mill; evaluating the faulty component of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter; wherein the first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature and inlet pressure of the main motor; the state classification model is a logistic regression model; the second characteristic parameter includes the first parameter of the main motor of the target coal mill, the second parameter of the main reducer, the third parameter of the grinding roller, the fourth parameter of the pressurizing device and the fifth parameter of the powder classifier; the first weight is predetermined based on the hierarchical analysis method.

[0212] When the computer program stored on the non-transitory computer-readable storage medium provided in the embodiment of the present application is executed, the above-mentioned coal mill failure risk assessment method is implemented. Its specific implementation method is consistent with the implementation method described in the embodiment of the aforementioned method and can achieve the same beneficial effects, which will not be repeated here.

[0213] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units. That is, they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0214] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus the necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or certain parts of the embodiment.

[0215] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A coal mill failure risk assessment method, characterized in that: include: Acquiring a first characteristic parameter of a target coal mill; determining the state of the target coal mill based on the first characteristic parameter and a state classification model; When the target coal mill is in a fault state, obtaining a second characteristic parameter of the target coal mill; evaluating a faulty component of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter; The first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature, and inlet pressure of the main motor; the state classification model is a logistic regression model; the second characteristic parameter includes a first parameter of the main motor of the target coal mill, a second parameter of the main reducer, a third parameter of the grinding roller, a fourth parameter of the pressurizing device, and a fifth parameter of the classifier; and the first weight is predetermined based on the analytic hierarchy process. The determining the state of the target coal mill based on the first characteristic parameter and the state classification model includes: Inputting the first feature parameter into the state classification model to obtain a first probability; When the first probability is greater than a probability threshold, determining the state of the target coal mill as a fault state; The evaluating the faulty component of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter includes: obtaining the health of each layer of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter; Based on the health status, a faulty component of the target coal mill is evaluated.

2. The coal mill failure risk assessment method according to claim 1, characterized in that: The obtaining of the health of each layer of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter includes: Obtaining a second weight corresponding to the health level of the current level based on the health level of the current level and the first weight corresponding to the health level of the current level; The health of the previous level is obtained based on the health of the current level and the second weight corresponding to the health of the current level.

3. The coal mill failure risk assessment method according to claim 2, characterized in that: The obtaining, based on the health level of the current level and the first weight corresponding to the health level of the current level, of the second weight corresponding to the health level of the current level includes: Based on each group of health degrees at the current level, a first weight corresponding to each group of health degrees, and a preset balancing function, a second weight corresponding to each group of health degrees is obtained.

4. The coal mill failure risk assessment method according to claim 3, characterized in that: The balance function is an exponential function.

5. A coal mill fault risk assessment device, characterized in that: include: A first acquisition module, configured to acquire a first characteristic parameter of a target coal mill; a classification module, configured to determine the state of the target coal mill based on the first characteristic parameter and a state classification model; a second acquisition module, configured to acquire a second characteristic parameter of the target coal mill when the target coal mill is in a fault state; an evaluation module, configured to evaluate a faulty component of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter; The first characteristic parameter includes at least one of the current, outlet pressure, outlet temperature, and inlet pressure of the main motor; the state classification model is a logistic regression model; the second characteristic parameter includes a first parameter of the main motor of the target coal mill, a second parameter of the main reducer, a third parameter of the grinding roller, a fourth parameter of the pressurizing device, and a fifth parameter of the classifier; and the first weight is predetermined based on the analytic hierarchy process. The determining the state of the target coal mill based on the first characteristic parameter and the state classification model includes: Inputting the first feature parameter into the state classification model to obtain a first probability; When the first probability is greater than a probability threshold, determining the state of the target coal mill as a fault state; The evaluating the faulty component of the target coal mill based on the second characteristic parameter and the first weight corresponding to the second characteristic parameter includes: obtaining the health of each layer of the target coal mill based on the second characteristic parameter and a first weight corresponding to the second characteristic parameter; Based on the health status, a faulty component of the target coal mill is evaluated.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the coal mill failure risk assessment method according to any one of claims 1 to 4 is implemented.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the coal mill failure risk assessment method according to any one of claims 1 to 4 is implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the coal mill failure risk assessment method according to any one of claims 1 to 4 is implemented.

Images