A coal mine equipment fault detection method, medium and system based on PHM

Abstract

The application provides a coal mine equipment fault detection method, medium and system based on PHM, belongs to the technical field of coal mine equipment fault detection, and comprises the following steps: collecting operation data based on PHM, including vibration, current and other indexes; next, the data is pretreated, denoised and features are extracted. Based on the pretreated data, a mathematical model covering dynamics, electrical characteristics and wear and tear is established, and an optimized model is obtained through parameter identification and optimization. The optimized model is discretized, a multi-dimensional state space is constructed, and normal state and fault state are defined. Based on the A* algorithm, the shortest path from the normal state to each fault state is searched, the difficulty of fault occurrence is calculated, and a fault weight vector is obtained. Real-time data is input into the optimized model, the distance from the current state to each fault endpoint is calculated, the fault weight is combined, and the fault probability distribution and early warning information are output, thereby solving the technical problem that the prior art cannot meet the requirements of fault diagnosis accuracy, reliability and explainability.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of coal mine equipment fault detection, and in particular relates to a coal mine equipment fault detection method, medium and system based on PHM. BACKGROUND

[0002] Due to the harsh environment of coal mines, equipment is subjected to long-term vibration impact, high temperature and humidity, and other harsh conditions, and is prone to various faults, such as bearing wear, gear damage, motor winding short circuit, etc. Once these faults occur, not only will the equipment be damaged, but also serious safety accidents may be caused. Therefore, how to effectively diagnose and warn the faults of coal mine equipment has become a key problem in the industry. At present, the commonly used methods for fault diagnosis of coal mine equipment mainly include: fault diagnosis based on experience rules, fault diagnosis based on data-driven, and fault diagnosis based on physical model.

[0003] 1. The fault diagnosis method based on experience rules mainly relies on the experience knowledge of experts, establishes a fault diagnosis rule base, and identifies faults through threshold value judgment of characteristic parameters. This method is simple and intuitive, but has limited scope of application and cannot cope with complex multi-fault situations.

[0004] 2. The fault diagnosis method based on data-driven uses machine learning, neural network and other technologies to extract fault features from a large amount of operation data and train a fault recognition model. This method can adapt to complex faults, but requires a large amount of historical data as support, and the model has weak diagnostic interpretability.

[0005] 3. The fault diagnosis method based on physical model establishes a mathematical model describing the operation mechanism of the equipment, and identifies faults by comparing the measured data with the model prediction results. This method has good interpretability, but often requires a large number of parameter tests and model identification, and the modeling process is complex.

[0006] In summary, the existing fault diagnosis methods have their own advantages and disadvantages, and it is difficult to meet the needs of fault diagnosis accuracy, reliability and interpretability in complex coal mine environments. Therefore, it is urgent to develop a new fault diagnosis method that can fully utilize equipment operation data and comprehensively apply physical models and data-driven technologies to achieve comprehensive diagnosis and prediction of coal mine equipment faults. SUMMARY

[0007] Therefore, the present application provides a coal mine equipment fault detection method, medium and system based on PHM, which can solve the technical problem that the prior art cannot meet the needs of fault diagnosis accuracy, reliability and interpretability in complex coal mine environments.

[0008] The present application is implemented as follows:

[0009] The first aspect of the application provides a coal mine equipment fault detection method based on PHM, wherein the method comprises the following steps:

[0010] S10, collecting operation data of the coal mine equipment, including vibration, temperature, pressure, current, speed, noise, oil analysis data, power consumption, equipment operation time, and environmental humidity;

[0011] S20, preprocessing the collected operation data, including denoising, normalization, and feature extraction;

[0012] S30, establishing a mathematical model of the coal mine equipment considering the operation data, including a set of dynamic equations, a set of electrical characteristic equations, a set of wear equations, and a set of fault experience equations;

[0013] S40, performing parameter identification and optimization on the mathematical model using the preprocessed data to obtain an optimized mathematical model;

[0014] S50, discretizing the optimized mathematical model to obtain a multi-dimensional state space, wherein each dimension represents a key parameter, a normal operation state is defined as a starting point, and different types of fault states are defined as end points;

[0015] S60, constructing a fault path network, regarding each possible fault evolution process as a path from the starting point to the end point, and regarding each node on the path as an intermediate state of the equipment;

[0016] S70, searching and optimizing each fault path using the A* algorithm, calculating the shortest path from the normal state to various fault states, and regarding the path length as the difficulty of fault occurrence; according to the lengths of the fault paths, calculating a fault weight vector, and regarding the fault weight as larger when the path is shorter;

[0017] S80, inputting real-time collected equipment operation data into the optimized mathematical model to obtain the position of the current equipment state in the state space, and calculating the distance from the position to each fault end point;

[0018] S90, combining the fault weight vector and the distance from the current state to each fault end point to calculate the final fault probability distribution, and outputting fault diagnosis results and warning information.

[0019] The set of dynamic equations is used to describe the mechanical motion characteristics of the equipment, including the relationship between displacement, angle, and angular velocity, and is specifically represented as follows:

[0020]

[0021] where m is the mass of the device (kg); x is the displacement (m); t is the time (s); c is the damping coefficient (N·s / m); k is the spring constant (N / m); F(t) is the external force (N); I is the moment of inertia (kg·m 2 ); θ is the angular displacement (rad); b is the angular damping coefficient (N·m·s / rad); τ s is the static friction torque (N·m); T(t) is the external torque (N·m); ω is the angular velocity (rad / s); α is the acceleration coefficient (rad / s 2 ); β is the velocity decay coefficient (1 / rad); ∈1, ∈2, ∈3 are error terms. Parameter acquisition method: m, I are obtained through device specification parameters; c, k, b, τ s are obtained through vibration test experiments, the steps including: 1) installing acceleration sensors at different positions of the device; 2) applying a known excitation force to the device; 3) recording the vibration response of the device; 4) fitting the vibration response curve using the least squares method to obtain the parameter values. F(t) and T(t) are measured in real time through force sensors and torque sensors. α and β are obtained through angular velocity measurement experiments, the steps including: 1) measuring the angular velocity of the device using a high-precision encoder; 2) recording the change of angular velocity over time under different working conditions; 3) fitting the angular velocity curve using a nonlinear regression method to obtain the parameter values.

[0022] Further, the electrical characteristic equation set is used to describe the electrical performance of the device, including the relationship between current, voltage, power, and electromagnetic torque, and is specifically represented as follows:

[0023]

[0024] P=VIcosφ+∈5;

[0025]

[0026] where L is the inductance (H); i is the current (A); R is the resistance (Ω); C is the capacitance (F); V(t) is the voltage (V); P is the power (W); I is the effective value current (A); φ is the power factor angle (rad); T e is the electromagnetic torque (N·m); τ is the electromagnetic time constant (s); K t is the torque constant (N·m / A); ∈4, ∈5, ∈6 are error terms. Parameter acquisition method: L, R, C are measured through an LCR tester. V(t) and i are measured in real time through a voltmeter and an ammeter. φ is measured through a power analyzer. τ and K t are obtained through motor parameter test experiments, the steps including: 1) applying a step current input to the motor; 2) recording the response curve of the electromagnetic torque; 3) fitting the response curve using system identification methods to obtain the parameter values.

[0027] Further, the wear equation set is used to describe the wear process of key components of the equipment, including the relationship between wear amount, material hardness, and remaining life, and is specifically expressed as follows:

[0028]

[0029] In the formula, W is the wear amount (mm); K w is the wear coefficient; F is the friction force (N); N is the number of contacts; H is the material hardness (HV); H0 is the initial hardness (HV); K h is the hardness attenuation coefficient; τ h is the hardness attenuation time constant (s); L is the remaining life (h); L0 is the initial expected life (h); W c is the critical wear amount (mm); ∈7, ∈8, ∈9 are error terms. Parameter acquisition method: K w is obtained through a wear test, and the steps include: 1) simulating the working conditions of the equipment on a test bench; 2) measuring the wear amount at different time points; 3) using a linear regression method to fit the wear curve to obtain K w . F is measured by a force sensor. N is recorded by a counter. H0, K h , τ h is obtained through a hardness test experiment, and the steps include: 1) periodically measuring the surface hardness of the component; 2) recording the change of the hardness with time; 3) using a nonlinear regression method to fit the hardness curve to obtain the parameter value. L0 and W c are determined according to the historical data of the equipment and experience values.

[0030] Further, the failure experience equation set is used to describe the evolution law of common failures of the equipment, including the relationship between failure probability, mean time between failures, and failure characteristic signal, and is specifically expressed as follows:

[0031]

[0032] In the formula, P f (t) is the failure probability; t is the running time (h); η is the scale parameter; β is the shape parameter; MTBF is the mean time between failures (h); λ1 is the basic failure rate (1 / h); λ2 is the degradation coefficient; γ is the degradation rate; S(t) is the failure characteristic signal strength; S0 is the initial signal strength; K s is the signal growth coefficient; T(τ) is the temperature function; f(T) is the temperature influence function; ∈ 10 , ∈ 11 , ∈ 12is the error term. Parameter acquisition method: η and β are obtained by statistical analysis of failure data, steps include: 1) collect equipment historical failure data; 2) use maximum likelihood estimation method to fit Weibull distribution, get parameter value. λ1, λ2 and γ are obtained by reliability test, steps include: 1) perform accelerated life test; 2) record the number of failures at different time points; 3) use nonlinear regression method to fit MTBF curve, get parameter value. S0, K s and f(T) are obtained by failure characteristic signal monitoring experiment, steps include: 1) monitor failure characteristic signal under different temperature conditions; 2) record the change of signal intensity with time and temperature; 3) use multivariate regression analysis to determine parameter value and temperature influence function form.

[0033] Further, the optimized mathematical model is discretized, that is, the continuous time mathematical model is converted into a discrete state space expression that can be used in computer programs; the discretization process is as follows:

[0034] (1) Select appropriate time step Δt, usually determined according to system dynamic characteristics and calculation accuracy requirements.

[0035] (2) Discretize the continuous time equation, for example, the displacement equation in the dynamic equation group:

[0036]

[0037] Where, x n represents the displacement of the nth time step, x′ n represents the velocity, which can be approximated by central difference:

[0038]

[0039] (3) Discretize other equation groups similarly.

[0040] (4) Define the state vector X = [x, x', θ, θ', ω, i, W, H, S] T , which contains the key variables of the system.

[0041] (5) Express the discretized equation group in state space form:

[0042] X n+1 = f(X n , U n ) + w n ;

[0043] Where f is a nonlinear state transition function, U n is the input vector, and w n is the process noise.

[0044] Further, the A* algorithm is used to search and optimize each fault path. In the process of calculating the shortest path from the normal state to various fault states, the construction of the fault path network is described as follows:

[0045] (1) Define the normal operating state as the starting point X0, and define different types of fault states as the terminal points X f，i , where i represents different fault types.

[0046] (2) Discretize the state space into a grid, and each grid point represents a possible intermediate state.

[0047] (3) Define the state transition cost function g(X i , X j ), which represents the cost from state X i to X j , which can be determined based on the state equation and physical constraints.

[0048] (4) Define the heuristic function h(X), which estimates the minimum cost from the current state X to the fault state, which can be designed based on the fault experience equation set.

[0049] (5) Use the A* algorithm to search for the shortest path from X0 to each X f，i :

[0050] a. Initialize the open list and the closed list;

[0051] b. Add the starting point X0 to the open list;

[0052] c. Loop until the target point is found or the open list is empty:

[0053] Select the node X c in the open list with the smallest f(X) = g(X) + h(X);

[0054] If X c is the target point, the path is found and the loop is exited;

[0055] Move X c from the open list to the closed list;

[0056] For each adjacent node X n of X c :

[0057] If X n is not in the closed list, calculate f(X n ) and add it to the open list;

[0058] If X n is already in the open list, update its f value if the new path is better;

[0059] (6) Repeat step (5) for each fault type to obtain all fault paths.

[0060] (7) Calculate the fault weight vector:

[0061] where L i is the length of the i-th fault path, w i is the corresponding weight.

[0062] Further, the operation data of the coal mine equipment is obtained from a PHM system of the coal mine or obtained by using sensors or instruments.

[0063] Further, the step of outputting the fault diagnosis result and the early warning information further comprises outputting the diagnosis result and the early warning information to the PHM system in a Json manner.

[0064] The second aspect of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores program instructions, and the program instructions are used to execute the PHM-based coal mine equipment fault detection method when the program instructions are run in a computer.

[0065] The third aspect of the present application provides a PHM-based coal mine equipment fault detection system, which comprises the computer readable storage medium.

[0066] Compared with the prior art, the PHM-based coal mine equipment fault detection method, medium and system have the following beneficial effects: 1. Comprehensive use of equipment operation data: the method collects rich operation data such as vibration, temperature, pressure, current, speed, noise and oil analysis of the equipment, which provides a solid foundation for subsequent analysis and modeling. Compared with a single fault feature, multi-dimensional operation data can better reflect the actual working state of the equipment.

[0067] 2. Establishment of a physical mechanism model: the method constructs a comprehensive mathematical model including a dynamic equation set, an electrical characteristic equation set, a wear equation set and a fault experience equation set, which can accurately describe the operation law and fault evolution mechanism of the equipment. Compared with a pure data-driven method, the physical model has stronger explainability and prediction ability.

[0068] 3. Optimization of model parameters and fault paths: the method uses parameter identification and model optimization techniques to make the mathematical model better fit the actual equipment characteristics. At the same time, the A* algorithm is used to search and optimize the fault path network to find the shortest path from the normal state to various fault states, which provides a quantitative basis for fault diagnosis. These methods greatly improve the accuracy and reliability of the overall diagnosis.

[0069] 4. Real-time monitoring of fault risk: The method inputs the real-time collected equipment operation data into the optimized mathematical model, timely calculates the distance from the current state to each fault endpoint, and gives the fault probability distribution combined with the fault path weight. In this way, not only can the current fault be found in time, but also the occurrence trend of future faults can be predicted, providing effective support for equipment maintenance.

[0070] In summary, the fault detection method of the present application fully integrates the advantages of physical modeling and data-driven, and realizes comprehensive diagnosis and prediction of coal mine equipment faults. Compared with the prior art, the method has significant improvement in diagnosis accuracy, reliability and interpretability, and solves the technical problems that the prior art cannot meet the needs of fault diagnosis accuracy, reliability and interpretability in complex coal mine environment. BRIEF DESCRIPTION OF DRAWINGS

[0071] Figure 1 The flowchart of the method provided by the present application. DETAILED DESCRIPTION

[0072] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application.

[0073] As Figure 1 shown, it is a flowchart of a coal mine equipment fault detection method based on PHM provided by the first aspect of the present application, and the method comprises the following steps:

[0074] S10, collecting operation data of the coal mine equipment, including vibration, temperature, pressure, current, speed, noise, oil analysis data, power consumption, equipment running time, and environmental humidity;

[0075] S20, pre-processing the collected operation data, including denoising, normalization and feature extraction;

[0076] S30, establishing a mathematical model of the coal mine equipment considering the operation data, including a set of dynamic equations, a set of electrical characteristic equations, a set of wear equations, and a set of fault experience equations;

[0077] S40, parameter identification and optimization of the mathematical model using the pre-processed data, to obtain an optimized mathematical model;

[0078] S50, discretizing the optimized mathematical model to obtain a multi-dimensional state space, wherein each dimension represents a key parameter, and defining the normal operation state as the starting point and different types of fault states as the endpoints;

[0079] S60, construct a fault path network, consider each possible fault evolution process as a path from the starting point to the ending point, and each node on the path represents an intermediate state of the device;

[0080] S70, search and optimize each fault path using the A* algorithm, calculate the shortest path from the normal state to each fault state, and the path length represents the difficulty of fault occurrence; according to the length of each fault path, calculate the fault weight vector, and the fault weight of the shorter path is larger;

[0081] S80, input the real-time collected device operation data into the optimized mathematical model to obtain the position of the current device state in the state space, and calculate the distance from the position to each fault endpoint;

[0082] S90, combine the fault weight vector and the distance from the current state to each fault endpoint to calculate the final fault probability distribution, and output the fault diagnosis result and warning information.

[0083] The specific implementation of the above steps is described in detail as follows:

[0084] The specific implementation of step S10 is to collect the operation data of the coal mine equipment. The purpose of this step is to use the PHM system or sensors to collect key parameters during the operation of the coal mine equipment, to provide basic data for subsequent mathematical modeling and fault diagnosis. Specifically, this step includes the following sub-steps:

[0085] Install various sensors: install vibration sensors, temperature sensors, pressure sensors, current sensors, speed sensors, noise sensors, etc. at key parts of the coal mine equipment to collect real-time operation data of the equipment. At the same time, install an environmental humidity sensor near the equipment. For important lubricated parts, an oil analyzer is also needed. These sensors should be selected from high-precision, high-temperature and high-humidity resistant industrial-grade devices.

[0086] Establish a data acquisition system: connect the above-mentioned sensors with industrial control devices to build a reliable data acquisition system. This system should have the functions of automatic acquisition, real-time storage, remote monitoring, etc. to ensure the integrity and continuity of various operation data. The sampling frequency should be reasonably selected according to the dynamic characteristics of the equipment, generally more than 100 Hz.

[0087] Calibrate the sensors: before formal use, the sensors need to be calibrated and calibrated to ensure the accuracy and reliability of the data. The calibration method includes using standard objects, comparative measurement, etc. The appropriate calibration method should be selected according to the type of sensor. After calibration, calibration check should be performed regularly to ensure data quality.

[0088] Real-time data acquisition: Once the data acquisition system is put into use, it starts to collect various operating parameter data of the equipment in real time, including vibration, temperature, pressure, current, speed, noise, oil analysis data, power consumption, equipment running time, environmental humidity, etc. These data will provide the basis for subsequent mathematical modeling and fault diagnosis.

[0089] Through the above sub-steps, the operating data of coal mine equipment can be comprehensively and accurately collected, laying the foundation for subsequent analysis and modeling. The key of this step lies in the selection of sensors, the reliability of the data acquisition system, and the continuity and accuracy of data acquisition.

[0090] Next is step S20, preprocessing the collected operating data:

[0091] The specific implementation of step S20 is to preprocess the collected operating data, including denoising, normalization and feature extraction. The purpose of this step is to eliminate noise interference in the data, so that the data has better signal-to-noise ratio, and at the same time extracts key features that are beneficial to fault diagnosis. It includes the following sub-steps:

[0092] Data denoising: filtering the collected raw data to remove high-frequency noise and interference generated during measurement. Methods such as Butterworth filter, wavelet transform, etc. can be used for noise suppression. The selection of filter parameters should be optimized according to the frequency spectrum characteristics of the signal to ensure that the signal distortion is not caused too much while denoising.

[0093] Data normalization: unifies different dimensional data to the [0, 1] interval to eliminate the influence of dimension. The minimum-maximum normalization formula can be used: where x n is the normalized data, x is the original data, x min and x max are the minimum and maximum values of the original data, respectively. This method can effectively eliminate the dimensional differences between different parameters.

[0094] Feature extraction: extracts key features that are beneficial to fault diagnosis from the preprocessed data. Common feature extraction methods include time domain features (mean, variance, peak factor, etc.), frequency domain features (power spectral density, peak frequency, etc.), time-frequency features (wavelet coefficients, Hilbert-Huang transform, etc.). Through comparative analysis, select features highly correlated with different fault modes.

[0095] Through the above data preprocessing steps, a set of data that has been denoised, normalized and feature extracted can be obtained, providing better input for subsequent mathematical modeling and fault diagnosis. The key of this step is to select appropriate filtering methods, normalization algorithms and feature extraction indicators, and fully exploit the fault diagnosis information contained in the data.

[0096] Next is step S30, establishing a mathematical model of the coal mine equipment:

[0097] The specific implementation of step S30 is to establish a mathematical model considering the operation data of the equipment, including a set of kinetic equations, a set of electrical characteristic equations, a set of wear equations, and a set of fault empirical equations. The purpose of this step is to build a mathematical description that can comprehensively describe the operating characteristics of the equipment, providing a theoretical basis for subsequent fault diagnosis and prediction. It includes the following sub-steps:

[0098] Establishing a set of kinetic equations: The set of kinetic equations is used to describe the mechanical motion characteristics of the equipment, including the relationship between key parameters such as displacement, angle, angular velocity, etc. It mainly includes two equations: one is the mass-spring-damper equation, which describes the linear motion of the equipment; the other is the moment of inertia-torque equation, which describes the rotational motion of the equipment. The parameters involved in these equations, such as mass, damping coefficient, and elastic coefficient, can be obtained through equipment specification parameters and experimental tests.

[0099] Establishing a set of electrical characteristic equations: The set of electrical characteristic equations is used to describe the electrical performance of the equipment, including the relationship between parameters such as current, voltage, power, and electromagnetic torque. It mainly includes three equations: one is the inductance-resistance-capacitance equation, which describes the dynamic characteristics of the circuit; the second is the active power equation, which describes the input power; the third is the electromagnetic torque equation, which describes the torque output of the motor. The parameters involved in these equations, such as inductance, resistance, and torque constant, can be obtained through test instruments and parameter identification experiments.

[0100] Establishing a set of wear equations: The set of wear equations is used to describe the wear process of key components of the equipment, including the relationship between parameters such as wear amount, material hardness, and remaining life. It mainly includes three equations: one is the wear rate equation, which describes the change of wear amount with time; the second is the material hardness equation, which describes the decay of hardness with time; the third is the remaining life equation, which describes the remaining usage time. The parameters involved in these equations, such as wear coefficient and hardness decay coefficient, can be obtained through wear tests and hardness tests.

[0101] Establishing a set of fault empirical equations: The set of fault empirical equations is used to describe the evolution law of common faults of the equipment, including the relationship between parameters such as fault probability, mean time between failures, and fault characteristic signals. It mainly includes three equations: one is the Weibull distribution function, which describes the change of fault probability with time; the second is the nonlinear MTBF equation, which describes the change of mean time between failures with running time; the third is the fault characteristic signal equation, which describes the change of signal intensity with time and temperature. The parameters involved in these equations, such as scale parameter and basic failure rate, can be obtained through reliability tests and signal monitoring experiments.

[0102] By establishing the above four types of equation sets, a comprehensive and detailed mathematical model is constructed, which can accurately describe the operating characteristics of the equipment, the evolution law of the fault, etc.

[0103] Next is step S40, parameter identification and optimization of the mathematical model:

[0104] The specific implementation of step S40 is to use the preprocessed data to identify and optimize the parameters of the mathematical model, and obtain the optimized mathematical model. The purpose of this step is to make the mathematical model better fit the operating characteristics of the actual equipment, and provide a more accurate basis for subsequent fault diagnosis and prediction. Specifically, the following sub-steps are included:

[0105] Model parameter identification: identify the unknown parameters in each equation of the mathematical model established in step S30. Parameter estimation algorithms such as least squares method and maximum likelihood estimation method can be used to identify and optimize the model parameters using preprocessed operating data. Through parameter identification, parameter values closer to the actual equipment characteristics can be obtained.

[0106] Model optimization: On the basis of parameter identification, further optimize the overall performance of the mathematical model. Intelligent optimization algorithms such as genetic algorithm and particle swarm optimization can be used to minimize the error between the model prediction value and the measured value, and obtain a more accurate model expression.

[0107] Model verification: apply the optimized mathematical model to new operating data for verification, check the prediction accuracy and robustness of the model. If the model performance cannot meet the requirements, the model structure can be adjusted appropriately or the parameter identification and optimization steps can be repeated until the model reaches the ideal state.

[0108] Through parameter identification and model optimization, a mathematical model closer to the actual equipment characteristics can be obtained. Such a model not only can accurately describe the operating rules of the equipment, but also can provide a more reliable theoretical basis for subsequent fault diagnosis and prediction. The key to this step is to select appropriate parameter identification and optimization algorithms, and to analyze the model verification results in depth.

[0109] Next is step S50, discretization of the optimized mathematical model:

[0110] The specific implementation of step S50 is to discretize the optimized mathematical model to obtain a multi-dimensional state space. The purpose of this step is to convert the continuous-time mathematical model into a discrete state space representation that can be used in computer programs, laying the foundation for subsequent fault path search. Specifically, the following sub-steps are included:

[0111] Time discretization: First, choose an appropriate time step Δt, usually determined by system dynamics and computational accuracy requirements. Then, discretize the continuous-time differential equations, for example, the displacement equation in the dynamic equation set:

[0112]

[0113] where x n is the displacement at the nth time step, x′ n is the velocity, which can be approximated by central difference. Similar discretization is performed for other equation sets.

[0114] State vector definition: Integrate the discretized equation set into a state space form, define the state vector X = [x, x', θ, θ', ω, i, W, H, S] T , which contains the key variables of the system. This way, the continuous-time mathematical model is converted into a discrete state space representation.

[0115] State transition equation establishment: Organize the discretized equation set into a state space form of the state transition equation:

[0116] X n+1 = f(X n , U n ) + W n ;

[0117] where f is a nonlinear state transition function, U n is the input vector, and w n is the process noise. This way, a multi-dimensional state space model for computer program is obtained.

[0118] Through the above discretization process, the original continuous-time mathematical model is converted into an operable discrete state space representation. This provides a basis for subsequent state space-based fault path search and fault diagnosis. The key to this step is to choose an appropriate time step, and whether the definition of the state vector can fully describe the key characteristics of the system.

[0119] Next is step S60, constructing a fault path network:

[0120] The specific implementation of step S60 is to construct a fault path network, considering each possible fault evolution process as a path from the normal state to the fault state. The purpose of this step is to establish an intuitive fault evolution model, providing a visual reference for subsequent fault diagnosis and prediction. This includes the following sub-steps:

[0121] Define state space: First, discretize the multi-dimensional state space established in step S50 into a grid, with each grid point representing a possible intermediate state. Define the normal operating state as the starting point X0 and different types of fault states as the end points X f，i , where i represents different fault types.

[0122] Determine state transition cost: Define a state transition cost function g(X i , X j ) representing the cost from state X i to X j . This cost can be determined based on the mathematical model and physical constraints established in step S30, reflecting the difficulty of transition between states.

[0123] Design heuristic function: Define a heuristic function h(X) for estimating the minimum cost from the current state X to the fault state. This function can be designed based on the fault experience equations in step S30, using parameters such as fault probability and remaining life for estimation.

[0124] Search for the shortest fault path: Use the A* algorithm to search and optimize each fault path, calculating the shortest path from the normal state to various fault states. The path length L i represents the difficulty of fault occurrence, with shorter paths corresponding to easier faults.

[0125] Calculate fault weight: According to the length L i of each fault path, calculate the fault weight vector The fault weight of the shorter path is greater, indicating that the fault is more likely to occur.

[0126] Through the above steps, a visual fault path network model is established. In this network, each path represents a possible fault evolution process, the path length reflects the difficulty of fault occurrence, and the weight represents the relative importance of each fault mode. Such a model provides an intuitive reference for subsequent fault diagnosis and prediction. The key lies in the reasonable definition of state transition cost and heuristic function, as well as the effective application of the A* algorithm.

[0127] Next is step S70, which uses the A* algorithm to optimize the fault path:

[0128] The specific implementation of step S70 is to use the A* algorithm to search and optimize each fault path, calculating the shortest path from the normal state to various fault states. The purpose of this step is to find the optimal evolution process of fault occurrence, providing a basis for subsequent fault diagnosis. It includes the following sub-steps:

[0129] Initialize open list and closed list: the open list stores nodes to be searched, and the closed list stores nodes that have been searched.

[0130] Add start node X0 to open list: start node X0 represents the normal operating state of the device.

[0131] Loop search until target node is found or open list is empty:

[0132] Select node X from open list with minimum f(X) = g(X) + h(X) c as current node. Here g(X) is the actual cost from start node to current node, and h(X) is the heuristic function estimate of the remaining cost.

[0133] If X c is target fault state X f，i , find shortest path and exit loop.

[0134] Move X c from open list to closed list.

[0135] For each adjacent node X c of X n :

[0136] If X n is not in closed list, calculate f(X n ) and add to open list.

[0137] If X n is already in open list, update its f value if new path is better.

[0138] Repeat step 3 for each fault type to get the shortest length L i of all fault paths.

[0139] Calculate fault weight vector: The shorter the path, the greater the fault weight, indicating that the fault is more likely to occur.

[0140] A* algorithm is an optimal path planning algorithm based on heuristic search, which selects the next search node by evaluating the total cost from the current node to the target node. This algorithm can effectively find the shortest path from the normal state to various fault states, providing reliable theoretical basis for subsequent fault diagnosis. The key to this step is to define the state transition cost function g(X) and heuristic function h(X) reasonably to improve the efficiency and accuracy of search.

[0141] Next, step S80, input real-time data into the optimized mathematical model:

[0142] The specific implementation of step S80 is to input the real-time collected equipment operation data into the optimized mathematical model to obtain the position of the current equipment state in the state space, and calculate the distance from the position to each fault endpoint. The purpose of this step is to use the optimized mathematical model to diagnose and analyze the real-time operation state of the equipment, and provide a basis for subsequent fault warning. Specifically, it includes the following sub-steps:

[0143] Real-time data input: input the real-time operation data of the equipment collected in step S10, such as vibration, temperature, current, etc., into the optimized mathematical model in step S40.

[0144] State space positioning: use the mathematical model to calculate the position X of the current equipment state in the multi-dimensional state space established in step S50 c . This position reflects the current operation state of the equipment.

[0145] Calculate the distance to the fault state: calculate the distance d i from the current state X c to each fault endpoint X f，i =||X c -X f，i ||. This distance reflects the closeness of the current state to various fault states.

[0146] Through this step, the running state of the equipment can be tracked in real time, and the closeness to various fault states can be evaluated. This provides key information for subsequent fault diagnosis and warning. The key of this step lies in the accuracy of the mathematical model and the reasonable design of the state space, ensuring accurate positioning of the current state and calculation of the distance to the fault state.

[0147] Finally, step S90 calculates the fault probability distribution and outputs the diagnosis result:

[0148] The specific implementation of step S90 is to combine the fault weight vector calculated in step S70 and the distance from the current state to each fault endpoint calculated in step S80 to calculate the final fault probability distribution, and output the fault diagnosis result and warning information. The purpose of this step is to comprehensively use the mathematical model and fault path network established in the previous steps to comprehensively diagnose and predict the current fault state of the equipment, and provide decision support for maintenance. Specifically, it includes the following sub-steps:

[0149] Calculate the fault probability distribution: according to the distance d i from the current state X c to each fault endpoint X f，i , and the fault weight w i calculated in step S70, calculate the occurrence probability of each fault mode:

[0150]

[0151] where a is a parameter representing the width of the probability distribution. This way, the probability distribution of each failure mode at the current time is obtained.

[0152] Output the diagnosis result: According to the calculated failure probability distribution, output the fault diagnosis result of the current equipment. Including:

[0153] the failure mode with the highest probability and its occurrence probability

[0154] the occurrence probability of other failure modes

[0155] the overall health status index of the current equipment

[0156] Provide failure warning: If the occurrence probability of a certain failure mode exceeds the preset threshold, timely issue a failure warning message to remind maintenance personnel to check and maintain. The threshold can be set according to actual experience, and it is generally recommended to be set between 0.2 and 0.3.

[0157] Through this step, the mathematical model and the failure path network established in the previous steps are comprehensively utilized to diagnose and predict the current failure state of the equipment, and a specific failure probability distribution is given to provide decision support for maintenance and repair. The key to this step is to reasonably design the failure probability calculation formula and reasonably set the failure warning threshold.

[0158] For step S90, the method of outputting the diagnosis result and warning information to the PHM system is as follows: convert the calculated failure probability distribution into a standard format (such as JSON or XML), define multiple warning thresholds, and store the results in a relational database. Use message queues to achieve real-time push, develop a web dashboard for visual display. At the same time, integrate multiple alarm notification methods, generate maintenance suggestions and operation guidance, provide historical query function, and integrate with enterprise asset management system. Finally, establish a feedback mechanism to record the comparison between actual failure conditions and prediction results, support continuous optimization of the model. This method can effectively realize the closed loop of equipment health management, and improve the accuracy and timeliness of fault detection.

[0159] In summary, this coal mine equipment fault detection method based on PHM makes full use of equipment operation data, and realizes comprehensive diagnosis and prediction of equipment failure through a series of steps such as mathematical modeling, failure path search, and real-time state monitoring.

[0160] In order to better understand and implement the present application, the following provides an embodiment of the method provided by the first aspect of the present application, and the details of each step in this embodiment are described as follows:

[0161] In step S10, the operation data of the coal mine equipment needs to be collected. Specifically, it includes:

[0162] 1) Install various sensors: Install vibration sensors, temperature sensors, pressure sensors, etc. at key parts of the equipment to collect operating parameters. Vibration sensors can measure displacement x(t) and angle θ(t) of the equipment, temperature sensors can measure equipment temperature T(t), and pressure sensors can measure equipment pressure p(t). At the same time, current sensors, speed sensors, noise sensors, etc. are also needed to collect electrical characteristics and environmental parameters. In addition, oil analyzers are also needed to measure the state of the equipment lubricating oil, etc. The installation of these sensors should meet the requirements of industrial environment, such as high temperature and high humidity resistance, etc.

[0163] 2) Establish a data acquisition system: Connect the above sensors with industrial control equipment to build a reliable data acquisition system. The system should have functions such as automatic acquisition, real-time storage, remote monitoring, etc. to ensure the integrity and continuity of various operating data. The sampling frequency should be reasonably selected according to the dynamic characteristics of the equipment, generally more than 100Hz to meet the needs of subsequent analysis.

[0164] 3) Calibrate the sensor: Before formal use, the sensors need to be calibrated and calibrated to ensure the accuracy and reliability of the data. Standard objects can be used for comparison and calibration, or input signals with known characteristics can be used for response calibration. After calibration, calibration checks should be made regularly to ensure data quality.

[0165] 4) Real-time data acquisition: Once the data acquisition system is put into use, it starts to collect various operating parameter data of the equipment in real time, including vibration displacement x(t), angle θ(t), temperature T(t), pressure p(t), current i(t), speed ω(t), noise n(t), oil analysis data, power consumption P(t), equipment running time t, environmental humidity H(t), etc. These data will provide a basis for subsequent mathematical modeling and fault diagnosis.

[0166] In step S20, the collected operating data need to be preprocessed. Specifically, it includes:

[0167] 1) Data denoising: Filter the original data to remove high-frequency noise interference in the measurement process. The Butterworth filter can be used, whose transfer function is:

[0168]

[0169] where s is the complex frequency, ω c is the cutoff frequency, and n is the filter order. By reasonably selecting ω c and n, high-frequency noise can be effectively suppressed while retaining valid signals.

[0170] 2) Data normalization: unify different dimensional data to the interval [0, 1] to eliminate the influence of dimension. The minimum-maximum normalization formula can be used:

[0171]

[0172] where x n is the normalized data, x is the original data, x min and x max are the minimum and maximum values of the original data, respectively. This method can effectively eliminate the dimensional differences between different parameters.

[0173] 3) Feature extraction: extract key features that are beneficial to fault diagnosis from preprocessed data. Common feature extraction methods include:

[0174] Time domain features:

[0175] Frequency domain features: f p = argmax f P xx (f);

[0176] Time-frequency features:

[0177] where μ is the mean, σ 2 is the variance, γ is the peak factor, P xx (f) is the power spectral density, f p is the peak frequency, and W(t, f) is the wavelet coefficient. By analyzing these features, pattern features related to different fault modes can be identified.

[0178] Through the above data preprocessing steps, a set of data after denoising, normalization and feature extraction is obtained, providing high-quality input for subsequent mathematical modeling and fault diagnosis.

[0179] In step S30, a mathematical model considering the equipment operation data needs to be established. Specifically, it includes:

[0180] 1) Dynamic equation set: the dynamic equation set is used to describe the mechanical motion characteristics of the equipment. For linear displacement motion, the mass-spring-damper equation can be established:

[0181]

[0182] where m is the equipment mass (kg), x is the displacement (m), t is the time (s), c is the damping coefficient (N·s / m), k is the elastic coefficient (N / m), F(t) is the external force (N), and ∈1 is the error term.

[0183] For rotational motion, a rotational inertia-torque equation can be established:

[0184]

[0185] where I is the rotational inertia (kg-m 2 ), θ is the angular displacement (rad), b is the angular damping coefficient (N-m-s / rad), τ s is the static friction torque (N-m), T(t) is the external torque (N-m), and ∈2 is the error term.

[0186] In addition, an angular velocity equation can be established:

[0187]

[0188] where ω is the angular velocity (rad / s), α is the acceleration coefficient (rad / s 2 ), β is the velocity decay coefficient (1 / rad), and ∈3 is the error term.

[0189] 2) Electrical characteristic equations: The electrical characteristic equations are used to describe the electrical performance of the device. An inductance-resistance-capacitance equation can be established:

[0190]

[0191] where L is the inductance (H), i is the current (A), R is the resistance (Ω), C is the capacitance (F), V(t) is the voltage (V), and ∈4 is the error term.

[0192] An active power equation can also be established:

[0193] P = VI cos φ + ∈5;

[0194] where P is the power (W), I is the effective value current (A), φ is the power factor angle (rad), and ∈5 is the error term.

[0195] and an electromagnetic torque equation:

[0196]

[0197] where T e is the electromagnetic torque (N-m), τ is the electromagnetic time constant (s), K t is the torque constant (N-m / A), and ∈6 is the error term.

[0198] 3) Wear equations: The wear equations are used to describe the wear process of the key components of the device. A wear rate equation can be established:

[0199]

[0200] where W is the wear amount (mm), K w is the wear coefficient, F is the friction force (N), N is the number of contacts, H is the material hardness (HV), ∈7is the error term.

[0201] At the same time, the hardness decay equation can also be established:

[0202]

[0203] where H0is the initial hardness (HV), K h is the hardness decay coefficient, τ h is the hardness decay time constant (s), ∈8is the error term.

[0204] and the remaining life equation:

[0205]

[0206] where L is the remaining life (h), L0is the initial expected life (h), W c is the critical wear amount (mm), ∈9is the error term.

[0207] 4) The failure experience equation set: The failure experience equation set is used to describe the evolution law of common failures of equipment. The failure probability equation can be established:

[0208]

[0209] where P f (t) is the failure probability, t is the running time (h), η is the scale parameter, β is the shape parameter, ∈ 10 is the error term.

[0210] At the same time, the mean time between failures (MTBF) equation can also be established:

[0211]

[0212] where λ1is the basic failure rate (1 / h), λ2is the degradation coefficient, γ is the degradation rate, ∈ 11 is the error term.

[0213] and the failure characteristic signal equation:

[0214]

[0215] where S(t) is the failure characteristic signal intensity, S0is the initial signal intensity, K s is the signal growth coefficient, T(τ) is the temperature function, f(T) is the temperature influence function, ∈ 12 is the error term.

[0216] By establishing the above four types of equations, a comprehensive and detailed mathematical model is constructed, which can accurately describe the operating characteristics of the equipment, the evolution law of the fault, etc. These equations and their parameters will provide strong support for subsequent fault diagnosis and prediction.

[0217] In step S40, parameter identification and optimization of the mathematical model are needed. Specifically, it includes:

[0218] 1) Model parameter identification: identify the unknown parameters in each equation of the mathematical model established in step S30. The least squares method can be used, that is, to find the parameter solution that minimizes the sum of squared residuals between predicted values and measured values:

[0219]

[0220] Where p is the parameter vector to be identified, y i is the ith measured value, is the ith predicted value. Using preprocessed operating data, the model parameters are identified and optimized to obtain parameter values closer to the actual equipment characteristics.

[0221] 2) Model optimization: on the basis of parameter identification, further optimize the overall performance of the mathematical model. Genetic algorithm can be used to iteratively optimize model parameters by simulating natural selection and genetic mechanisms:

[0222] p k+1 = p k + α k d k ;

[0223]

[0224] Where p k is the parameter vector of the kth iteration, α k and β k are step parameters, is the gradient of the objective function. By minimizing the error between the model predicted values and the measured values, a more accurate model expression is obtained.

[0225] 3) Model verification: apply the optimized mathematical model to new operating data for verification, check the prediction accuracy and robustness of the model. If the model performance cannot meet the requirements, the model structure can be adjusted or the parameter identification and optimization steps can be repeated until the model reaches the ideal state.

[0226] Through the above parameter identification and model optimization, a mathematical model closer to the actual equipment characteristics can be obtained, providing a more reliable theoretical basis for subsequent fault diagnosis and prediction.

[0227] In step S50, the optimized mathematical model needs to be discretized. Specifically, it includes:

[0228] 1) Time discretization: Select an appropriate time step Δt, usually determined according to system dynamic characteristics and calculation accuracy requirements. Then discretize the continuous-time differential equation by difference, taking the displacement equation in the dynamic equation set as an example:

[0229]

[0230] where x n is the displacement at the nth time step, x′ n is the velocity, which can be approximated by central difference:

[0231]

[0232] 2) State vector definition: Integrate the discretized equation set into a state space form, define the state vector X = [x, x', θ, θ', ω, i, W, H, S] T , which contains the key variables of the system.

[0233] 3) State transition equation establishment: Organize the discretized equation set into a state space form of state transition equation:

[0234] X n+1 = f(X n , U n ) + w n ;

[0235] where f is a nonlinear state transition function, U n is the input vector, and w n is the process noise.

[0236] Through the above discretization, the original continuous-time mathematical model is converted into an operable discrete state space representation, providing a basis for subsequent state space-based fault path search and fault diagnosis.

[0237] In step S60, the fault path network needs to be constructed. Specifically, it includes:

[0238] 1) Define the state space: First, discretize the multi-dimensional state space established in step S50 into a grid, and each grid point represents a possible intermediate state. Define the normal operating state as the starting point X0, and define different types of fault states as the terminal points X i , where i represents different fault types.

[0239] 2) Determine the state transition cost: Define the state transition cost function g(X j , X i), represents the cost from state X i to X j . This cost can be determined based on the mathematical model and physical constraints established in step S30, reflecting the difficulty of transition between states. For example:

[0240]

[0241] where ||·|| is the L2 norm, and h(X) is the heuristic function.

[0242] 3) Design the heuristic function: Define the heuristic function h(X) to estimate the minimum cost from the current state X to the failure state. It can be designed based on the failure experience equation set in step S30, using parameters such as failure probability and remaining life to estimate:

[0243]

[0244] where P f (t) is the failure probability, and L(X, X f , i) is the distance from state X to the failure endpoint X f，i .

[0245] 4) Search for the shortest failure path: Use the A* algorithm to search and optimize each failure path, and calculate the shortest path from the normal state to various failure states. The path length L i represents the difficulty of failure, and the shorter the path, the easier the failure occurs.

[0246] 5) Calculate the failure weight: According to the length L i of each failure path, calculate the failure weight vector:

[0247]

[0248] The shorter the path, the greater the failure weight, indicating that the failure is more likely to occur.

[0249] Through the above steps, a visual failure path network model is established. In this network, each path represents a possible failure evolution process, the path length reflects the difficulty of failure, and the weight represents the relative importance of each failure mode. Such a model provides an intuitive reference for subsequent failure diagnosis and prediction.

[0250] In step S70, the A* algorithm is used to optimize the failure path. Specifically, it includes:

[0251] 1) Initialize the open list and the closed list: The open list stores the nodes to be searched, and the closed list stores the nodes that have been searched.

[0252] 2) Add start node X0 to open list: start node X0 represents the normal operating state of the device.

[0253] 3) Loop search until target node is found or open list is empty:

[0254] a. Select node X from open list with minimum f(X) = g(X) + h(X) c as current node. Here g(X) is the actual cost from start node to current node, and h(X) is the heuristic function estimate of the remaining cost.

[0255] b. If X c is the target fault state X f，i , find the shortest path and exit the loop.

[0256] c. Move X c from open list to closed list.

[0257] d. For each adjacent node X c of X n :

[0258] If X n is not in the closed list, calculate f(X n ) and add it to the open list;

[0259] If X n is already in the open list, update its f value if the new path is better;

[0260] 4) Repeat step 3 for each fault type to get the shortest length L i of all fault paths.

[0261] 5) Calculate fault weight vector:

[0262]

[0263] The shorter the path, the greater the fault weight, indicating that the fault is more likely to occur.

[0264] A* algorithm is an optimal path planning algorithm based on heuristic search, which selects the next search node by evaluating the total cost from the current node to the target node. This algorithm can effectively find the shortest path from the normal state to various fault states, providing a reliable theoretical basis for subsequent fault diagnosis.

[0265] In step S80, real-time data needs to be input into the optimized mathematical model. Specifically, it includes:

[0266] 1) Real-time data input: The real-time running data of the equipment collected in step S10, such as vibration displacement x(t), angle θ(t), temperature T(t), current i(t), etc., are input into the mathematical model optimized in step S40.

[0267] 2) State space positioning: Using the mathematical model, the position X of the current equipment state in the multi-dimensional state space established in step S50 is calculated c . This position reflects the current running state of the equipment, which can be expressed as:

[0268] X c =f -1 (X n , U n );

[0269] Where f is the state transition function, X n is the current state vector, and U n is the current input vector.

[0270] 3) Calculate the distance to the fault state: Calculate the distance d i from the current state X c to each type of fault end X f，i , which can use L2 norm:

[0271] d i =||X c -X f，i ||2;

[0272] This distance reflects the closeness of the current state to various fault states.

[0273] Through this step, the running state of the equipment can be tracked in real time, and its closeness to various fault states can be evaluated. This provides key information for subsequent fault diagnosis and early warning.

[0274] In step S90, the fault probability distribution needs to be calculated and the diagnosis result is output. Specifically including:

[0275] 1) Calculate the fault probability distribution: According to the distance d i from the current state X c to each fault end X f，i , and the fault weight w i calculated in step S70, the occurrence probability of each fault mode is calculated:

[0276]

[0277] Where σ is a parameter representing the width of the probability distribution, which can be set according to the actual situation. In this way, the probability distribution of each fault mode at the current time is obtained.

[0278] 2) Output the diagnosis result: According to the calculated failure probability distribution, output the failure diagnosis result of the current device. Including:

[0279] a. The failure mode with the highest probability and its occurrence probability

[0280] b. The occurrence probability of other failure modes

[0281] c. The overall health status index of the device at present, which can be defined as h = 1 - ∑ i P f，i ;

[0282] 3) Provide failure warning: If the occurrence probability P f，i of a certain failure mode exceeds the preset threshold P t , timely issue a failure warning information to remind the maintenance personnel to check and maintain. The threshold P t can be set according to actual experience, and it is generally recommended to be set between 0.2 and 0.3.

[0283] Through this step, the mathematical model and the failure path network established in the foregoing are comprehensively utilized to comprehensively diagnose and predict the current failure state of the device, and a specific failure probability distribution is given to provide decision support for maintenance.

[0284] The second aspect of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores program instructions, and the program instructions are used to execute the above-mentioned coal mine equipment failure detection method based on PHM when running in a computer; the computer readable storage medium can be a U disk or a computer hard disk. The third aspect of the present application provides a coal mine equipment failure detection system based on PHM, which comprises the above-mentioned computer readable storage medium; at the same time, the system also connects the PHM platform through API, supports obtaining data or feedback from the PHM platform; the system can also be used as a micro-service source of PHM, and is used for detecting the failure of the device by the PHM.

[0285] Specifically, the principle of the present application is: a PHM technical framework is adopted, and multiple advanced technical means such as physical model analysis, data-driven modeling and real-time monitoring are comprehensively applied. The specific principle is as follows:

[0286] 1. Physical model analysis: this method first establishes a set of mathematical models describing the operation mechanism of the device, including dynamic equation sets, electrical characteristic equation sets, wear equation sets and failure experience equation sets. These physical models can accurately reflect the working characteristics of each subsystem of the device, and provide a theoretical basis for subsequent state evaluation and failure diagnosis. Compared with the pure data-driven method, the physical model has stronger explainability and generalization ability.

[0287] 2. Data-driven modeling: Based on the physical model, this method uses the measured equipment operation data to identify parameters and optimize the model, making the mathematical model better fit the actual equipment characteristics. This fusion of physical modeling and data-driven methods takes full advantage of both techniques, while preserving interpretability and improving overall model accuracy.

[0288] 3. Fault path search: Based on the optimized mathematical model, this method defines the normal operation state and various fault states of the equipment as nodes in the state space, and uses the A* algorithm to search for the shortest path from the normal state to each fault state. The path length reflects the difficulty of fault occurrence, providing a quantitative basis for diagnosis. This state-space-based fault path analysis method can intuitively demonstrate the mechanism of fault evolution and provide a theoretical basis for fault prediction.

[0289] 4. Real-time monitoring and early warning: This method inputs real-time collected equipment operation data into the optimized mathematical model, calculates the current state's position in the state space in real time, and evaluates its distance to each fault endpoint. Combined with the fault path weights obtained in step 3, the occurrence probability distribution of each fault mode can be calculated. Once a certain fault probability exceeds the preset threshold, the system immediately issues a warning message to provide decision support for maintenance. This real-time monitoring and early warning mechanism can effectively prevent equipment failure.

[0290] In summary, the fault detection method of the present invention fully integrates the advantages of physical modeling and data-driven methods, and uses advanced technical means in key links such as establishing a mathematical model describing the mechanism of the equipment, optimizing model parameters, searching for fault evolution paths, and real-time monitoring of fault risks. This comprehensive and systematic PHM technical framework enables the method to overcome the limitations of existing technologies and truly achieve accurate diagnosis and effective prediction of coal mine equipment faults.

[0291] In order to better understand and implement the present invention, an example of a specific application scenario of the present invention is provided below: A coal mining enterprise frequently encounters coal mine equipment failure problems during production, resulting in production interruptions and seriously affecting production efficiency. To solve this problem, the enterprise decides to use the PHM-based coal mine equipment fault detection method proposed by the present invention to conduct comprehensive monitoring and fault diagnosis of key equipment.

[0292] 1. Data collection

[0293] First, the enterprise installs various sensors on the main coal mine equipment, as shown in Table 1:

[0294] Table 1 Sensor installation

[0295] Sensor type Mounting location Measured parameter Vibration sensor Motor bearing, gearbox Displacement x(t), angle Θ(t) Temperature sensor Motor winding, gearbox Temperature T(t) Pressure sensor Hydraulic oil line Pressure p(t) Current sensor Motor input Current i(t) Speed sensor Motor shaft Angular velocity ω(t) Noise sensor Around the equipment Noise n(t) Oil analysis Gearbox, hydraulic system Oil condition

[0296] These sensors are all industrial-grade high-precision devices that can withstand the harsh conditions of coal mine environments. In addition, the company has installed environmental humidity sensors near the equipment to monitor the working environment.

[0297] To build a reliable data acquisition system, the company uses industrial control devices to connect the above sensors, achieving automatic acquisition, real-time storage, and remote monitoring functions. The system's sampling frequency is set to 200Hz to ensure that it can capture the key dynamic characteristics of the equipment in operation.

[0298] Before being put into use, the company calibrates and calibrates each sensor. Taking the vibration sensor as an example, first, it is calibrated at different frequencies by a standard exciter to obtain the frequency response characteristics of the sensor. Then, the response curve of the sensor is recorded during the operation of the equipment, and the least squares method is used to fit the damping coefficient c and the elastic coefficient k. Similar calibration methods are also applied to other sensors to ensure the accuracy and reliability of the data.

[0299] After the above steps, the company has established a complete data acquisition system that can comprehensively and accurately monitor the running state of coal mine equipment.

[0300] 2. Mathematical modeling

[0301] Based on the collected equipment operation data, the company then establishes a mathematical model that describes the mechanism of the equipment, including a set of dynamic equations, a set of electrical characteristic equations, a set of wear equations, and a set of fault experience equations.

[0302] (1) Dynamic equations

[0303] For linear displacement motion of the equipment, a mass-spring-damper equation can be established:

[0304]

[0305] Where, the equipment mass m = 500 kg, the damping coefficient c = 5 × 10 4 N·s / m, the elastic coefficient k = 2 × 10 6 N / m, and the external force F(t) is measured by the force sensor in real time. By fitting the vibration response of the equipment, the company determines the above dynamic parameters.

[0306] For rotational motion, a rotational inertia-torque equation can be established:

[0307]

[0308] Where, the rotational inertia I = 50 kg·m 2 , the angular damping coefficient b = 100 N·m·s / rad, and the static friction torque τs =500 N·m, the external torque T(t) is measured in real time by a torque sensor. These parameter values ​​are also obtained by fitting the response curve.

[0309] In addition, the angular velocity equation can be established:

[0310]

[0311] Wherein, the acceleration coefficient α = 10 rad / s 2 The velocity decay coefficient β = 0.05 / rad. These parameters were obtained through angular velocity testing experiments.

[0312] (2) Electrical characteristic equations

[0313] For the dynamic characteristics of a circuit, the inductor-resistor-capacitor equation can be established:

[0314]

[0315] In this model, the inductance L = 0.05H, the resistance R = 1Ω, the capacitance C = 0.01F, and the voltage V(t) is measured in real time by a voltmeter. These parameters are obtained by measuring an LCR meter.

[0316] For active power, a power equation can be established:

[0317] P = VIcosφ + ∈5;

[0318] Among them, the effective current I = 50A and the power factor angle φ = 0.8rad were obtained by measuring with a power analyzer.

[0319] For electromagnetic torque, a torque equation can be established:

[0320]

[0321] Wherein, the electromagnetic time constant τ = 0.05s, and the torque constant K t = 1 N·m / A. These parameters were obtained through motor parameter testing experiments.

[0322] (3) Wear equations

[0323] For the wear process, a wear rate equation can be established:

[0324]

[0325] Among them, the wear coefficient K w =1×10 -6 mm 2F = 500 N, the number of contacts N was recorded by a counter, and the material hardness H = 300 HV. These parameters were obtained by the wear test.

[0326] The hardness decay can be described by the following equation:

[0327]

[0328] where the hardness decay coefficient K h = 50 HV, and the decay time constant τ h = 10 4 s. These parameters were obtained by the hardness test.

[0329] Finally, the remaining life can be expressed by the following equation:

[0330]

[0331] where the initial expected life L0= 10 4 h, and the critical wear W c = 1 mm. These parameters were determined based on the equipment historical data and experience.

[0332] (4) Failure empirical equation set

[0333] For the failure probability, the Weibull distribution equation can be used to describe:

[0334]

[0335] where the scale parameter η = 1000 1 / h, and the shape parameter β = 1.5 1 / h. These parameters were obtained by statistical analysis of the historical failure data of the equipment.

[0336] The mean time between failures (MTBF) can be described by the following nonlinear equation:

[0337]

[0338] where the basic failure rate λ1= 0.001 1 / h, the degradation coefficient λ2= 0.0001 1 / h, and the degradation rate γ = 0.00011 / h. These parameters were obtained by accelerated life testing.

[0339] The failure characteristic signal strength can be expressed by the following equation:

[0340]

[0341] where the initial signal strength S0= 50, and the signal growth coefficient K s= 0.1, temperature impact function f(T) = 1 + 0.01T. These parameters are obtained from the fault signature monitoring experiment.

[0342] In summary, the enterprise has established a comprehensive mathematical model that can better describe the dynamic characteristics, electrical characteristics, wear process, and fault evolution of the equipment. This provides a theoretical basis for subsequent fault diagnosis.

[0343] 3. Model optimization

[0344] In order to make the mathematical model better reflect the characteristics of the actual equipment, the enterprise uses parameter identification and model optimization techniques to optimize it.

[0345] First, the enterprise uses the pre-processed operating data to identify the unknown parameters in each equation using the least squares method. Taking the displacement equation in the dynamic equation set as an example, the vibration response curve is fitted using the following formula to obtain the optimized m, c, and k values:

[0346]

[0347] Similar parameter identification methods are also applied to other equation sets to make the model parameters more realistic.

[0348] Next, the enterprise uses genetic algorithms to optimize the overall model. The objective function is to minimize the error between the predicted and measured values, and the parameters of each equation are iteratively optimized to obtain a more accurate mathematical description. After convergence, the optimized model is applied to new operating data for verification, and it is found that the prediction accuracy and robustness of the model have been significantly improved.

[0349] After the above optimization steps, the enterprise has established a mathematical model that can accurately describe the actual characteristics of the equipment, laying the foundation for subsequent fault diagnosis.

[0350] 4. Fault path analysis

[0351] Based on the optimized mathematical model, the enterprise next constructs a fault path network to analyze the evolution of the equipment from a normal state to various fault states.

[0352] First, the enterprise discretizes the multi-dimensional state space established in step 2 into a grid, with each grid point representing a possible intermediate state. The normal operating state of the equipment is defined as the starting point X0, and different types of fault states such as gear damage, bearing wear, and motor failure are defined as the end points X f,i .

[0353] Then, the enterprise defines the state transition cost function g(X i ,X j ), which represents the cost of transitioning from state X i to Xj difficulty of transition. This cost function takes into account the physical constraints in the mathematical model, such as:

[0354]

[0355] where ||X j -X i || is the distance between states, P f (t j ) is the failure probability at state X j , L(X j , X f,i ) is the distance from state X j to the failure end X f,i .

[0356] The company also designs a heuristic function h(X) to estimate the minimum cost from the current state to each type of failure end:

[0357]

[0358] With the state transition cost and heuristic function, the company uses the A* algorithm to search and optimize each failure path, calculating the shortest path from the normal state to each failure state. The path length L i represents the difficulty of failure occurrence, and the shorter the path, the easier the failure occurs.

[0359] Finally, according to the length L i of each failure path, the company calculates the failure weight vector:

[0360]

[0361] The shorter the path, the greater the failure weight, indicating that this failure mode is more worthy of attention.

[0362] Through the above steps, the company establishes a visual failure path network model, providing a theoretical basis for subsequent fault diagnosis.

[0363] 5. Real-time monitoring and early warning

[0364] With the optimized mathematical model and failure path analysis results, the company begins real-time monitoring and failure warning of the equipment.

[0365] First, the company inputs the real-time operating data collected in step 1, such as vibration displacement x(t), angle θ(t), temperature T(t), current i(t), etc., into the optimized mathematical model. The model is used to calculate the current equipment state position X c in the multi-dimensional state space:

[0366] Xc = f -1 (X n , U n );

[0367] where f is the state transition function, X n is the current state vector, and U n is the current input vector.

[0368] Next, the enterprise calculates the distances d i from the current state X c to each failure endpoint X f，i :

[0369] d i = ||X c - X f，i ||2;

[0370] These distances reflect the closeness of the current state to various failure states.

[0371] Finally, the enterprise combines the failure weight vectors w i calculated in step 4 to calculate the probability distribution of each failure mode:

[0372]

[0373] where σ = 0.1 is the width parameter of the probability distribution.

[0374] According to the calculation results, the enterprise outputs the failure diagnosis report of the current device, including:

[0375] The most probable failure mode is gear breakage, with a probability of 28.5%.

[0376] The probabilities of other failure modes are: bearing wear 18.2%, motor failure 12.6%, and pipeline leakage 8.7%.

[0377] The overall health status index of the device is 0.82, indicating that the overall working condition of the device is good.

[0378] However, if the probability of a certain failure mode exceeds 30%, the enterprise will immediately issue a failure warning message to remind maintenance personnel to check and maintain. This real-time monitoring and warning mechanism can help the enterprise discover potential equipment hazards in time and effectively prevent the occurrence of failures.

[0379] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any skilled person in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A PHM-based coal mine equipment fault detection method, characterized in that, The method comprises the following steps: S10, collecting operation data of the coal mine equipment, including vibration, temperature, pressure, current, speed, noise, oil analysis data, power consumption, equipment operation time, and environmental humidity; S20, preprocessing the collected operation data, including denoising, normalization, and feature extraction; S30, establishing a mathematical model of the coal mine equipment considering the operation data, including a set of dynamic equations, a set of electrical characteristic equations, a set of wear equations, and a set of fault experience equations; S40, performing parameter identification and optimization on the mathematical model using the preprocessed data to obtain an optimized mathematical model; S50, discretizing the optimized mathematical model to obtain a multi-dimensional state space, wherein each dimension represents a key parameter, a normal operation state is defined as a starting point, and different types of fault states are defined as end points; S60, constructing a fault path network, regarding each possible fault evolution process as a path from the starting point to the end point, and regarding each node on the path as an intermediate state of the equipment; S70, searching and optimizing each fault path using an A* algorithm to calculate the shortest path from the normal state to various fault states, and regarding the path length as the difficulty of fault occurrence; According to the lengths of the fault paths, a fault weight vector is calculated, and the fault weight is greater for a shorter path. S80, inputting the real-time collected equipment operation data into the optimized mathematical model to obtain the position of the current equipment state in the state space, and calculating the distances from the position to the fault end points; S90, combining the fault weight vector and the distances from the current state to the fault end points to calculate the final fault probability distribution, and outputting fault diagnosis results and warning information.

2. The PHM-based coal mine equipment fault detection method according to claim 1, characterized in that, The set of dynamic equations is used to describe the mechanical motion characteristics of the equipment, including the relationships among displacement, angle, and angular velocity.

3. The PHM-based coal mine equipment fault detection method of claim 2, wherein, The set of electrical characteristic equations is used to describe the electrical performance of the equipment, including the relationships among current, voltage, power, and electromagnetic torque.

4. The PHM-based coal mine equipment fault detection method of claim 3, wherein, The set of wear equations is used to describe the wear process of key components of the equipment, including the relationships among wear amount, material hardness, and remaining life.

5. The PHM-based coal mine equipment fault detection method of claim 4, wherein, The set of fault experience equations is used to describe the evolution law of common faults of the equipment, including the relationships among fault probability, mean time to failure, and fault characteristic signals.

6. The PHM-based coal mine equipment fault detection method of claim 5, wherein, Discretizing the optimized mathematical model specifically converts the continuous-time mathematical model into a discrete state space representation that can be used in computer programs.

7. The PHM-based coal mine equipment fault detection method of claim 6, wherein, The operation data of the coal mine equipment is obtained from a PHM system of the coal mine or obtained using sensors or instruments.

8. The PHM-based coal mine equipment fault detection method of claim 7, wherein, The step of outputting fault diagnosis results and warning information further comprises outputting the fault diagnosis results and warning information to the PHM system in the form of Json.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program instructions, which, when executed in a computer, are used to perform the coal mine equipment fault detection method based on PHM according to any one of claims 1-8.

10. A PHM based coal mine equipment fault detection system, characterized in that, The computer-readable storage medium according to claim 9. The computer-readable storage medium according to claim 9.

Images