Bearing bush reliability calculation and residual life prediction method and device, equipment and medium

Abstract

The invention relates to the technical field of rotating machinery bearing bush vibration, and discloses a bearing bush reliability calculation and residual life prediction method, device, equipment and medium, and the method comprises the steps: obtaining an equipment state output vector of fault equipment and a corresponding judgment specification vector when a bearing bush has a fragmentation fault; judging whether the equipment state output vector and the judgment specification vector conform to normal distribution or not, and directly establishing a limit state equation based on a judgment result, or establishing a limit state equation based on a preset limit checking point; calculating a reliability index of a preset life cycle of the equipment based on the limit state equation, and calculating a corresponding failure probability; and establishing an equipment residual life distribution function based on the failure probability, and performing curve fitting on the equipment residual life distribution function to obtain the equipment residual service life. The problem that the reliability and the residual life of the fault sliding bearing of the steam turbine generator unit are difficult to accurately calculate is solved.

Description

Technical Field

[0001] This invention relates to the field of rotating machinery bearing vibration technology, specifically to methods, devices, equipment, and media for bearing reliability calculation and remaining life prediction. Background Technology

[0002] The actual wear mechanism of bearing pairs is very complex. Frequent start-up and shutdown of the unit and bearing removal measurement tests are not permitted on-site. Furthermore, damaged bearings often undergo structural changes, leading to overlapping and mutually excitation of the rotor-bearing kinematic pairs, exhibiting coupled random characteristics. Therefore, in practice, it is difficult to accurately calculate the reliability and remaining life of faulty sliding bearings in steam turbine generator sets using dynamic theoretical models. When a unit fails, due to factors such as grid requirements, immediate shutdown is not possible. Therefore, it is essential to quickly assess the structural reliability of the bearings and accurately predict the remaining safe life to monitor the fault evolution process and prevent major safety accidents. Summary of the Invention

[0003] In view of this, the present invention provides a method, apparatus, equipment and medium for calculating bearing reliability and predicting remaining life, so as to solve the problem that it is difficult to accurately calculate the reliability and remaining life of faulty sliding bearings in steam turbine generator sets in the prior art.

[0004] In a first aspect, the present invention provides a method for calculating bearing reliability and predicting remaining life, the method comprising:

[0005] When a bearing bush breaks, the equipment status output vector of the faulty equipment and the corresponding evaluation standard vector are obtained.

[0006] Determine whether the device status output vector and the corresponding evaluation standard vector conform to a normal distribution, and establish the limit state equation directly based on the judgment result, or establish the limit state equation based on the preset limit verification point.

[0007] The reliability index of the equipment's preset life cycle is calculated based on the limit state equation, and the corresponding failure probability is calculated based on the reliability index.

[0008] The remaining service life of the equipment is obtained by establishing a distribution function of the remaining service life based on the failure probability and then performing curve fitting on the distribution function.

[0009] This invention provides a method for calculating bearing reliability and predicting remaining life. When a bearing experiences a fracture failure, considering that the generator set must continue operating for a period of time to complete specific tasks of the power grid, an algorithm for predicting bearing reliability and remaining life based on the Joint-Curve Method (JC method) is designed, using the information state equation as the theoretical basis. First, a limit state equation based on an arbitrary distribution of random response is established. Then, the reliability and comprehensive failure probability of the fault complete vector are calculated in different time periods. Finally, the remaining life prediction result is obtained through curve fitting, solving the safety assessment problem of continued operation after a failure. By judging the distribution type (normal / non-normal) of the equipment state output vector and the evaluation standard vector, the limit state equation can be flexibly established directly or based on preset limit verification points, improving the adaptability to complex data distributions. By curve fitting the remaining life distribution function, a quantitative and accurate prediction of the remaining service life of the equipment is achieved, providing a quantitative basis for generator set operation decisions after a failure. This solves the problem in existing technologies of accurately calculating the reliability and remaining life of faulty sliding bearings in steam turbine generator sets.

[0010] In an optional implementation, the bearing reliability calculation and remaining life prediction method further includes: determining whether the bearing has experienced a breakage failure by means of:

[0011] Obtain the time-domain waveform of the equipment bearing vibration, and extract the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics of the bearing vibration time-domain waveform;

[0012] When the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics are distorted, the bearing bush is determined to have broken.

[0013] This invention provides a method for calculating bearing reliability and predicting remaining life. It proposes using the nonlinear random characteristics of bearing vibration time-domain waveform and the distortion of its wide-spectrum amplitude-frequency characteristics as the basis for judging bearing breakage failure. This method clarifies the specific characteristic indicators for fault identification and improves the accuracy and pertinence of fault judgment.

[0014] In one optional implementation, the device status output vector of the faulty device and the corresponding evaluation criterion vector are obtained, including:

[0015] Based on the preset information state equation, the output vector corresponding to the specified state of the faulty equipment is extracted to obtain the equipment state output vector and the corresponding evaluation standard vector; the evaluation standard is a comprehensive evaluation space formed based on relevant standards, design data, and operating experience of similar equipment or relevant expert experience.

[0016] This invention provides a method for calculating bearing reliability and predicting remaining life. It proposes extracting output vectors corresponding to specified states of faulty equipment based on a preset information state equation. This provides clear theoretical support for obtaining equipment state output vectors and evaluation standard vectors, ensuring the scientific and standardized nature of vector extraction. The evaluation standard is defined as an evaluation space formed by integrating relevant standards, design data, operational experience of similar equipment, and expert experience. This clarifies the basis for the composition of the evaluation standard vectors, improving the objectivity and accuracy of reliability calculation and life prediction.

[0017] In one optional implementation, the limit state equation is established directly based on the judgment result, or the limit state equation is established based on a preset limit verification point, including:

[0018] When the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, the limit state equation is directly established based on the device state output vector and the corresponding evaluation standard vector.

[0019] When the device state output vector and the corresponding evaluation standard vector do not conform to the normal distribution, the device state output vector and the corresponding evaluation standard vector are subjected to equivalent normal processing based on the preset limit verification points to obtain a normally distributed random vector, and the limit state equation is established based on the normally distributed random vector.

[0020] This invention provides a method for calculating bearing reliability and predicting remaining life. It addresses the distribution types (normal / non-normal) of the equipment state output vector and the evaluation standard vector, employing either direct establishment or equivalent normalization processing to establish limit state equations. This enhances the adaptability of the equations to different data distributions and ensures the accuracy of reliability calculations. Furthermore, it clarifies the operational path for equivalent normalization processing based on preset limit verification points when the vector does not conform to a normal distribution, providing a standardized method for non-normal data conversion and guaranteeing the uniformity and scientific rigor of subsequent reliability index calculations.

[0021] In one optional implementation, the device state output vector and the corresponding evaluation specification vector are subjected to equivalent normalization based on preset limit verification points to obtain a normally distributed random vector, including:

[0022] At the preset limit verification points, calculate the distribution function and probability density of the non-normally distributed device state output vector and the corresponding evaluation standard vector;

[0023] Based on the corresponding distribution function and probability density, the standard normal distribution function is used to perform equivalent normalization transformation to obtain a normally distributed random vector.

[0024] This invention provides a method for calculating bearing reliability and predicting remaining life. It clarifies that at a preset limit verification point, the distribution function and probability density of a non-normally distributed vector are calculated, and then transformed using a standard normal distribution function. The specific steps of equivalent normalization are refined to ensure the feasibility and accuracy of the conversion from non-normal to normal vectors, thus improving the completeness of the method's technical details. A clear logical chain is established for calculating the distribution function / probability density and transforming to the standard normal function, providing a clear theoretical basis and operational standard for the normalization of non-normally distributed vectors. This ensures the reliability of the transformation results and lays an accurate data foundation for the subsequent establishment of the limit state equation.

[0025] In one optional implementation, the reliability index of the device within a preset lifespan is calculated based on the limit state equation, including:

[0026] The equipment's preset lifespan is divided into multiple analysis periods;

[0027] When the device status output vector and the corresponding evaluation standard vector for each analysis period conform to a normal distribution, a standard transformation is performed on the limit state equation for each analysis period to obtain the standard limit state surface for each analysis period. The shortest distance from the origin of the preset dimension space standard normal coordinate system to the standard limit state surface is used as the reliability index within the preset life cycle of the device.

[0028] When the device status output vector and the corresponding evaluation standard vector for each analysis period do not conform to a normal distribution, the coordinates of the preset limit verification point are initialized, the direction cosine sum is calculated based on the preset limit verification point coordinates, and the limit verification point coordinates are updated using the direction cosine sum; the updated limit verification point coordinates are substituted into the limit state equation for each analysis period to iteratively calculate the device reliability, and the reliability of all analysis periods is combined to obtain the reliability index.

[0029] This invention provides a method for calculating bearing reliability and predicting remaining life. By dividing the preset lifespan into multiple analysis periods, it considers the time-varying characteristics of the output parameters of the faulty equipment deteriorating over time, enabling the reliability index to dynamically reflect the safety status of the equipment at different stages, and the calculation results are more consistent with actual operating conditions. Through techniques such as limit state surface equations, direction cosines, and iterative updates, and strictly adhering to the mathematical logic of engineering structural reliability theory and the JC method, the calculation process of the reliability index is ensured to be scientifically controllable, and the accuracy of the results meets the needs of engineering decision-making. Combining the reliability of each period to form a full lifespan reliability index provides a phased safety assessment basis for scenarios where equipment needs to continue operating after failure, facilitating maintenance personnel to formulate targeted operating strategies.

[0030] In one optional implementation, a remaining lifespan distribution function is established based on the failure probability, and curve fitting is performed on the remaining lifespan distribution function to obtain the remaining lifespan of the equipment, including:

[0031] A failure distribution function is established based on the failure probability, and the failure distribution function is transformed into a remaining life distribution function of the equipment based on the conditional probability formula.

[0032] A curve fitting tool was used to fit the distribution function of the remaining life of the equipment, and the remaining life of the equipment was calculated based on the curve fitting.

[0033] This invention provides a method for calculating bearing reliability and predicting remaining life. Based on the formulas for failure probability and conditional probability, it derives the remaining life distribution function, strictly adhering to probability and statistics theory to ensure that the distribution function accurately characterizes the random nature of the remaining life, providing a reliable mathematical model for subsequent predictions. By fitting the distribution function with a curve fitting tool, the abstract probability distribution is transformed into a directly readable remaining life value, achieving a leap from qualitative analysis to quantitative prediction. This provides a precise time reference for power grid task planning after unit failures, improving the scientific nature of operation and maintenance decisions. Directly linking the failure probability with the remaining life distribution function creates a closed loop between reliability index calculation, failure probability analysis, and remaining life prediction, ensuring logical consistency between the life prediction results and previous reliability assessments, thus enhancing the systematicity and coherence of the overall method.

[0034] Secondly, the present invention provides a bearing reliability calculation and remaining life prediction device, the device comprising:

[0035] The vector acquisition module is used to acquire the equipment status output vector and the corresponding evaluation standard vector of the faulty equipment when the bearing bush breaks.

[0036] The limit state equation establishment module is used to determine whether the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, and to directly establish the limit state equation based on the judgment result, or to establish the limit state equation based on the preset limit verification point.

[0037] The reliability and failure probability calculation module is used to calculate the reliability index of the equipment's preset life cycle based on the limit state equation, and to calculate the corresponding failure probability based on the reliability index.

[0038] The equipment remaining useful life prediction module is used to establish a distribution function of equipment remaining useful life based on the failure probability, and to perform curve fitting on the distribution function of equipment remaining useful life to obtain the equipment remaining useful life.

[0039] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the bearing reliability calculation and remaining life prediction method of the first aspect or any corresponding embodiment described above.

[0040] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the bearing reliability calculation and remaining life prediction method of the first aspect or any corresponding embodiment described above.

[0041] Fifthly, the present invention provides a computer program product, including computer instructions, which are used to cause a computer to execute the bearing reliability calculation and remaining life prediction method of the first aspect or any corresponding embodiment described above. Attached Figure Description

[0042] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating the bearing reliability calculation and remaining life prediction method according to an embodiment of the present invention.

[0044] Figure 2 This is a flowchart illustrating another method for calculating bearing reliability and predicting remaining life according to an embodiment of the present invention.

[0045] Figure 3This is a flowchart illustrating another method for calculating bearing reliability and predicting remaining life according to an embodiment of the present invention.

[0046] Figure 4 This is a schematic diagram of the function and limit verification point based on the bearing normal distribution response according to an embodiment of the present invention;

[0047] Figure 5 This is a graph showing the relationship between failure probability and reliability according to an embodiment of the present invention;

[0048] Figure 6 This is a simplified diagram of the shaft system structure according to an embodiment of the present invention;

[0049] Figure 7 This is a spectrum diagram of the #11 bearing according to an embodiment of the present invention;

[0050] Figure 8 This is a spectrum diagram of the #12 bearing according to an embodiment of the present invention;

[0051] Figure 9 This is a time-domain waveform diagram (low-frequency range) of the #11 bearing according to an embodiment of the present invention;

[0052] Figure 10 This is a time-domain waveform diagram (high-frequency range) of the #11 bearing according to an embodiment of the present invention;

[0053] Figure 11 This is a time-domain waveform diagram (low-frequency range) of the #12 bearing according to an embodiment of the present invention;

[0054] Figure 12 This is a time-domain waveform diagram (high-frequency range) of the #12 bearing according to an embodiment of the present invention;

[0055] Figure 13 This is a spectrum comparison chart (low frequency range) of the #11 bearing at 12-hour intervals from March 23rd to 24th according to an embodiment of the present invention;

[0056] Figure 14 This is a waveform comparison chart (low frequency range) of the #11 bearing at 12-hour intervals from March 23rd to 24th, according to an embodiment of the present invention.

[0057] in, Figures 7 to 14 The three boxes in the middle, from top to bottom, represent the axial, vertical, and horizontal directions, respectively;

[0058] Figure 15 These are measured vibration diagrams of the #11 bearing in various directions according to an embodiment of the present invention;

[0059] Figure 16 This is a block diagram illustrating the principle of the time-varying reliability sample analysis method for bearing #11 according to an embodiment of the present invention.

[0060] Figure 17 This is a schematic diagram of the time-varying reliability trend and remaining life prediction of bearing #11 according to an embodiment of the present invention;

[0061] Figure 18 This is a structural block diagram of the bearing reliability calculation and remaining life prediction device according to an embodiment of the present invention;

[0062] Figure 19 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0063] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] Currently, in practice, it is difficult to accurately calculate the reliability and remaining life of faulty sliding bearings in steam turbine generator sets using dynamic theoretical models. In engineering, situations where equipment must continue operating with faults until a certain purpose is achieved due to insurmountable objective reasons are not uncommon, such as a fighter jet malfunctioning and needing to make a safe emergency landing. This invention provides a method for calculating bearing reliability and predicting remaining life. Considering that the unit must continue operating for a period of time to complete specific tasks of the power grid, and based on the information state equation, a bearing reliability and remaining life prediction algorithm based on the JC method (verification point method) is implemented. This achieves a more convenient, faster, and more effective analysis and calculation of equipment reliability and remaining life, solving the problem of accurately calculating the reliability and remaining life of faulty sliding bearings in steam turbine generator sets in existing technologies.

[0065] According to an embodiment of the present invention, a method for calculating bearing reliability and predicting remaining life is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0066] This embodiment provides a method for calculating bearing reliability and predicting remaining life, which can be used in sliding bearing systems. Figure 1 This is a flowchart of the bearing reliability calculation and remaining life prediction method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps:

[0067] Step S101: When the bearing bush breaks, obtain the equipment status output vector of the faulty equipment and the corresponding evaluation standard vector.

[0068] Specifically, bearing shell breakage refers to structural damage in mechanical equipment (especially rotating machinery such as steam turbines, generators, and water pumps) where the bearing shell component supporting the rotation of the shaft system experiences partial or complete fracture, cracking, or breakage due to material fatigue, overload, lubrication failure, installation misalignment, or foreign object intrusion. This type of failure is a serious mechanical failure that directly disrupts the normal operation of the shaft system and can even lead to equipment shutdown, cascading failures, or safety accidents. For example, the failure manifests as obvious cracks on the bearing shell surface and fragments falling off. In severe cases, the bearing shell base may completely shatter, possibly accompanied by secondary damage such as journal wear and shaft misalignment. During bearing shell operation, equipment vibration increases sharply (especially radial vibration), abnormal noise (such as metallic impact sounds), and the bearing shell temperature rises rapidly (due to lubrication failure or increased friction). Metal debris may also be mixed into the lubricating oil.

[0069] The equipment status output vector refers to a set of multi-dimensional physical parameters collected by sensors, monitoring systems, and other means after a bearing failure, characterizing the current operating state of the faulty equipment. Its core function is to quantitatively reflect the dynamic response level of the equipment under fault conditions, including: vibration characteristic parameters (such as effective value, peak value, and kurtosis of vibration acceleration); temperature parameters (such as bearing surface temperature and lubricating oil temperature); and oil parameters (such as the concentration of wear particles in the oil and the viscosity change rate). The equipment status output vector is a dynamic, time-varying vector, and its value gradually changes as the equipment operates for longer periods and the fault worsens.

[0070] The evaluation standard vector is a set of multi-dimensional benchmark parameters corresponding to the equipment status output vector, used to determine whether the equipment status output vector is within the safe operating range. Its core function is to provide quantitative evaluation criteria, sourced from: industry standards or specifications (such as bearing vibration limit standards, temperature safety thresholds); equipment design data (such as the allowable parameter range for fault conditions provided by the manufacturer); operating experience with similar equipment (such as critical values ​​derived from historical fault data statistics); and expert experience (customized evaluation thresholds for specific operating conditions). The evaluation standard vector can be dynamically adjusted according to the equipment's operational needs.

[0071] Sensors (such as vibration acceleration sensors, infrared temperature sensors, and online oil monitoring sensors) are installed at key locations on the faulty equipment to ensure that the collected data accurately reflects the bearing failure status. The collected raw data is filtered (to remove noise), normalized (to unify dimensions), and features are extracted (such as kurtosis and peak factor extracted from the vibration waveform) to form a structured equipment status output vector X(t).

[0072] By reviewing the industry standards and equipment design manuals related to bearing bushes, as well as the historical safe operating parameter ranges of similar units under similar faults, an initial evaluation dataset is formed. Combined with the special requirements of the current power grid task on the unit operation (such as needing to operate continuously for 48 hours), the initial specifications are dynamically adjusted. The adjusted evaluation standards of each parameter are arranged in the order of the dimensions of the equipment status output vector to form a one-to-one corresponding evaluation specification vector S(t).

[0073] Step S102: Determine whether the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, and directly establish the limit state equation based on the judgment result, or establish the limit state equation based on the preset limit verification point.

[0074] Specifically, the preset limit verification point is, in engineering reliability analysis, the geometric point that represents the shortest distance from the origin of the standard normal coordinate system to the limit state surface. The normal distribution, also known as the Gaussian distribution, is usually denoted as X ~ N(μ, σ). 2 ).

[0075] Where μ is the expected value (mean) of the normal distribution, σ 2 It is the variance of the normal distribution. A normal distribution with μ = 0 and σ = 1 is called the standard normal distribution.

[0076] Determine whether the equipment state output vector X(t) and the evaluation standard vector S(t) after the bearing breakage failure conform to a normal distribution. If they conform to a normal distribution, then the limit state equation is directly established based on the judgment result. If they do not conform to a normal distribution, then the limit state equation is established based on the preset limit verification point.

[0077] Step S103: Calculate the reliability index of the equipment's preset life cycle based on the limit state equation, and calculate the corresponding failure probability based on the reliability index.

[0078] Specifically, the reliability index is the shortest distance from the origin of the standard normal coordinate system to the limit state surface, reflecting the safety margin of the equipment.

[0079] A preset lifespan refers to a time period pre-set during the design, manufacturing, operation, maintenance, or analysis of equipment, based on factors such as usage requirements, design standards, industry specifications, and expected functional goals. This period measures the estimated time it will take for the equipment to complete its specified functions or achieve its expected performance indicators under normal operating conditions.

[0080] For example, a certain type of industrial motor might have its preset lifespan set to 10 years during the design phase, based on its material strength, operating load, and maintenance plan. Similarly, a precision instrument might have its preset lifespan set to 5000 hours based on the wear characteristics of its core components and the rate of accuracy degradation. Preset lifespans provide a baseline time framework for equipment reliability analysis, remaining lifespan assessment, and maintenance strategy development, serving as an important reference for conducting full lifecycle management and performance evaluation of equipment.

[0081] If the device state output vector and the corresponding evaluation criterion vector conform to a normal distribution, the limit state equation is standardized to obtain the limit state surface equation. According to the principle of engineering structural reliability, in 2n-dimensional space, the shortest distance from the origin of this standard normal coordinate system to the limit state surface is... This refers to the reliability index β of the equipment structure.

[0082] If the device status output vector and the corresponding evaluation standard vector do not conform to the normal distribution, the reliability index β of the device's preset life cycle is obtained by formula derivation based on the preset limit verification point.

[0083] When the equipment state parameter vector and its corresponding evaluation criterion vector conform to a normal distribution, its limit state equation Z (i.e., the normal distribution function) also conforms to a normal distribution. There is a precise mathematical relationship between the failure probability of the limit state equation Z and its reliability index β, such as... Figure 5 As shown, there is a one-to-one correspondence between the failure probability of the equipment and the reliability index β. The failure probability is obtained by calculating β.

[0084] Step S104: Establish the equipment remaining life distribution function based on the failure probability, and perform curve fitting on the equipment remaining life distribution function to obtain the equipment remaining service life.

[0085] Specifically, the remaining lifespan T of a device that has been operating safely for t hours under fault-free conditions, and can continue to operate until failure or malfunction occurs, is called the remaining lifespan T of the device with a service life of t. t .

[0086] A failure distribution function for equipment lifetime is established based on failure probability, which represents the probability of failure when the operating time is less than or equal to t. Given that the equipment has operated for t hours without failure, the remaining lifetime T is... t The distribution function is derived from the failure distribution function.

[0087] If the distribution pattern of the remaining lifespan of the equipment is known, fitting curves can be generated using fitting tools such as MATLAB to determine the remaining lifespan of the equipment.

[0088] This embodiment provides a method for calculating bearing reliability and predicting remaining life. When a bearing experiences a breakage failure, considering that the generator set must continue operating for a period of time to complete specific tasks for the power grid, a bearing reliability and remaining life prediction algorithm based on the Joint-Curve Method (JC method) is designed, using the information state equation as the theoretical basis. First, a limit state equation based on an arbitrary distribution of random response is established. Then, the reliability and comprehensive failure probability of the fault complete vector are calculated in different time periods. Finally, the remaining life prediction result is obtained through curve fitting, solving the safety assessment problem of continued operation after a failure. By judging the distribution type (normal / non-normal) of the equipment state output vector and the evaluation standard vector, the limit state equation can be flexibly established directly or based on preset limit verification points, improving the adaptability to complex data distributions. By curve fitting the remaining life distribution function, a quantitative and accurate prediction of the remaining service life of the equipment is achieved, providing a quantitative basis for generator set operation decisions after a failure. This solves the problem in existing technologies of accurately calculating the reliability and remaining life of faulty sliding bearings in steam turbine generator sets.

[0089] This embodiment provides a method for calculating bearing reliability and predicting remaining life, which can be used in sliding bearing systems. Figure 2 This is a flowchart of the bearing reliability calculation and remaining life prediction method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:

[0090] Step S201: Obtain the time-domain waveform of the equipment bearing vibration, and extract the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics of the bearing vibration time-domain waveform; when the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics are distorted, it is determined that the bearing bush has broken.

[0091] Specifically, the bearing vibration time-domain waveform is the original signal reflecting the bearing's operating state. The bearing vibration time-domain waveform can be obtained by installing sensors at radial or axial vibration sensitive points on the bearing housing.

[0092] When the bearing is operating normally, the nonlinear characteristics of the vibration signal are weak and the randomness is stable; during a fracture failure, due to the increased metal impact and friction, the nonlinearity and randomness are significantly enhanced. The nonlinear random characteristics of the vibration time-domain waveform can be represented by a probability density function.

[0093] Broadband amplitude-frequency characteristics reflect the energy distribution of vibration signals across different frequency bands. When a bearing breaks, it can cause energy anomalies across a wide frequency band. Key indicators of broadband amplitude-frequency characteristics include, but are not limited to, the proportion of energy at characteristic frequencies, the wideband energy distribution, and spectral peak values.

[0094] When the probability density function, characteristic frequency energy ratio, broadband energy distribution, and spectral peak value are distorted, the bearing bush is determined to have broken.

[0095] Step S202: When the bearing bush breaks, obtain the equipment status output vector of the faulty equipment and the corresponding evaluation standard vector.

[0096] Specifically, step S202 includes:

[0097] Step a: Based on the preset information state equation, extract the output vector corresponding to the specified state of the faulty equipment to obtain the equipment state output vector and the corresponding evaluation standard vector; the evaluation standard is an evaluation space formed by comprehensively considering relevant standards, design data, and operating experience of similar equipment or relevant expert experience.

[0098] Specifically, in the scenario of bearing failure analysis, the preset information state equation is a mathematical model that describes the dynamic relationship between the input, state and output of the faulty equipment. Its core is to associate the operating conditions (input) and internal state (such as bearing wear degree and stress distribution) of the equipment with observable output parameters (such as vibration and temperature) through quantification equations, so as to provide a theoretical framework for extracting the equipment state output vector.

[0099] The formula for the information state equation is as follows:

[0100]

[0101] Where x(t) represents the system state vector (e.g., unmeasurable responses such as bearing clearance, bearing offset, deflection, bearing cracking, dynamic and static rubbing, oil film instability, etc.); ρ(t) represents the system source input vector (e.g., speed, active power, reactive power, valve sequence and opening, vacuum, lubricating oil temperature, jacking oil pressure, etc.); y(t) represents the system output vector (e.g., measurable responses such as vertical, horizontal, and axial vibrations of various parts of the rotor and bearing housing, bearing temperature and return oil temperature, etc.); A(t) represents the coefficient matrix, which represents the influence factor of the system's inherent characteristics between the current state and the next state; B1(t) represents the feedback input matrix, which represents the degree of influence of the output response on the system state; if the feedback can be ignored, this matrix can be zero. B2(t) represents the source input influence matrix, indicating the degree of influence of each source input on the system state; C(t) represents the output matrix, indicating the influence factor between the output and the system state; D(t) represents the inertia matrix, indicating the direct effect of the source input on the output (Note: when the two are not directly related, D(t) = 0).

[0102] Current theories and calculation methods for equipment reliability analysis mostly assess safety margins by establishing the difference between the equipment's dynamic state and the dynamic limit of the vector under investigation. Since any standard is based on the dynamic limits of materials such as strength and stiffness, response standards or specifications can be directly used instead of vector limits.

[0103] Assume that X is a set of equipment state output vectors corresponding to a certain state of the equipment, found based on the above information state equation. The measured equipment state output vector at any time t is X(t), and the corresponding evaluation standard is S(t). The latter is an evaluation space formed by comprehensively considering X(t) based on relevant standards or design data, as well as the operating experience of similar equipment or the experience of relevant experts.

[0104] When the device state output vector is X(t) = [X1(t), X2(t), ..., X... n The corresponding evaluation norm vector is S(t) = [S1(t), S2(t), ..., S(t)]. n (t)].

[0105] Step S203: Determine whether the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, and directly establish the limit state equation based on the judgment result, or establish the limit state equation based on the preset limit verification point.

[0106] Specifically, step S203 includes:

[0107] Step S2031: When the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, the limit state equation is directly established based on the device state output vector and the corresponding evaluation standard vector.

[0108] Specifically, when the device state output vector X(t) = [X1(t), X2(t), ..., X... n [(t)] and the corresponding evaluation norm vector S(t) = [S1(t), S2(t), ..., S... n When the two are independent and satisfy a normal distribution, the limit state equation (also called the limit state surface equation) is: Z(t)=g[X(t),S(t)]=0, where Z(t) is the value of the limit state equation.

[0109] For X i S i (For ease of expression, (t) is omitted below.) After standard transformation, the formula is as follows:

[0110]

[0111] The limit state equation then transforms into:

[0112]

[0113] in, Let μ represent the variances of the normal distributions of the device state output vector X(t) and the corresponding evaluation standard vector S(t), respectively; Xi μ Si Let X(t) and S(t) represent the mean values ​​of the normal distributions of the device status output vector X(t) and the corresponding evaluation criterion vector S(t), respectively.

[0114] Step S2032: When the device state output vector and the corresponding evaluation standard vector do not conform to the normal distribution, the device state output vector and the corresponding evaluation standard vector are subjected to equivalent normal processing based on the preset limit verification point to obtain a normally distributed random vector, and the limit state equation is established based on the normally distributed random vector.

[0115] Specifically, the limit verification point P is designed in the original coordinate system (referring to the coordinate system where the device status output vector is located). * The coordinates are:

[0116]

[0117] in,

[0118]

[0119] Obviously:

[0120] g(X * ,S * )=0(6);

[0121] When X(t) and S(t) do not conform to a normal distribution, a verification point P needs to be designed. * The two are converted into a normalized random vector (X′,S′) by equivalent normalization, and then the above formula (6) is used for calculation.

[0122] in, These represent the direction angles corresponding to the device status output vector and the evaluation standard vector, respectively, and β represents the reliability index.

[0123] In an optional implementation, step S2032 includes:

[0124] Step b: At the preset limit verification points, calculate the distribution function and probability density of the non-normally distributed device state output vector and the corresponding evaluation standard vector; based on the corresponding distribution function and probability density, perform equivalent normalization transformation using the standard normal distribution function to obtain a normally distributed random vector.

[0125] Specifically, when X(t) and S(t) do not conform to a normal distribution, a verification point P needs to be designed.* After performing equivalent normalization, the two are transformed into a normally distributed random vector (X′, S′) using the following formula:

[0126]

[0127] Where X′(t) and S′(t) are X(t) and S(t) at the design verification point P, respectively. * [X * (t),S * The equivalent normalized variable at [t] is the derivative of the corresponding cumulative distribution function; F(·) and f(·) are the cumulative distribution function and the probability density function, respectively; Φ(·) is the standard normal function. Let μ be the probability density function of the standard normal distribution, and μ(·) and σ(·) be the mean and standard deviation, respectively.

[0128] The limit state equation is then transformed into a normal distribution form:

[0129] Z′=g[X′,S′]=0 (11).

[0130] The bearing reliability calculation and remaining life prediction method provided in this embodiment clearly defines the transformation process at a preset limit check point. This involves calculating the distribution function and probability density of a non-normally distributed vector and then converting it using the standard normal distribution function. The specific steps of equivalent normalization are refined to ensure the feasibility and accuracy of the conversion from non-normal to normal vectors, thus improving the completeness of the method's technical details. A clear logical chain is established for the calculation of the distribution function / probability density and the transformation using the standard normal function, providing a clear theoretical basis and operational standards for the normalization of non-normally distributed vectors. This ensures the reliability of the transformation results and lays an accurate data foundation for the subsequent establishment of the limit state equation.

[0131] Step S204: Calculate the reliability index for the preset lifespan of the equipment based on the limit state equation, and calculate the corresponding failure probability based on the reliability index. For details, please refer to [link to relevant documentation]. Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0132] Step S205: Establish a remaining life distribution function for the equipment based on the failure probability, and perform curve fitting on the remaining life distribution function to obtain the remaining service life of the equipment. For details, please refer to [link to relevant documentation]. Figure 1 Step S104 of the illustrated embodiment will not be described again here.

[0133] The bearing reliability calculation and remaining life prediction method provided in this embodiment proposes using the nonlinear random characteristics of the bearing vibration time-domain waveform and the distortion of its broad-spectrum amplitude-frequency characteristics as the basis for judging bearing breakage failures. This clarifies the specific characteristic indicators for fault identification, improving the accuracy and specificity of fault judgment. It proposes extracting the output vector corresponding to the specified state of the faulty equipment based on a preset information state equation, providing clear theoretical support for obtaining the equipment state output vector and the evaluation standard vector, ensuring the scientific and standardized nature of vector extraction. The evaluation standard is defined as an evaluation space formed by comprehensively considering relevant standards, design data, operating experience of similar equipment, and expert experience, clarifying the basis for the composition of the evaluation standard vector and improving the objectivity and accuracy of reliability calculation and life prediction.

[0134] This embodiment provides a method for calculating bearing reliability and predicting remaining life, which can be used in sliding bearing systems. Figure 3 This is a flowchart of the bearing reliability calculation and remaining life prediction method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps:

[0135] Step S301: When a bearing bush fractures, obtain the equipment status output vector and the corresponding evaluation standard vector of the faulty equipment. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0136] Step S302: Determine whether the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, and directly establish the limit state equation based on the determination result, or establish the limit state equation based on preset limit verification points. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0137] Step S303: Calculate the reliability index of the equipment's preset life cycle based on the limit state equation, and calculate the corresponding failure probability based on the reliability index.

[0138] Specifically, step S303 includes:

[0139] Step S3031: Divide the preset life cycle of the equipment into multiple analysis periods.

[0140] Specifically, after the bearing breaks, the rate of deterioration of the equipment status output vector (such as vibration and temperature) is non-linear with time (it may deteriorate slowly in the early stage, and deteriorate more rapidly in the later stage due to crack propagation). The time period division needs to be able to capture key deterioration nodes (such as dividing a time period for every 10% deterioration of the reliability index).

[0141] Step S3032: When the device status output vector and the corresponding evaluation standard vector for each analysis period conform to a normal distribution, perform a standard transformation on the limit state equation for each analysis period to obtain the standard limit state surface for each analysis period. Use the shortest distance from the origin of the preset dimension space standard normal coordinate system to the standard limit state surface as the reliability index within the preset life cycle of the device.

[0142] Specifically, based on formulas (3) and (4), the limit state equations for each analysis period are transformed using a standard transformation to obtain the standard limit state surface for each analysis period. The preset dimension space is set to 2n-dimensional space. According to the principle of engineering structural reliability, in the above 2n-dimensional space, the shortest distance from the origin of the standard normal coordinate system to the limit state surface is... This refers to the reliability index β of the equipment structure.

[0143] Step S3033: When the device status output vector and the corresponding evaluation standard vector for each analysis period do not conform to the normal distribution, initialize the preset limit verification point coordinates, calculate the direction cosine sum based on the preset limit verification point coordinates, update the limit verification point coordinates using the direction cosine sum; substitute the updated limit verification point coordinates into the limit state equation for each analysis period to iteratively calculate the device reliability, and combine the reliability of all analysis periods to obtain the reliability index.

[0144] Specifically, the response levels of various output parameters of the faulty structural components gradually deteriorate over time; therefore, the reliability of the faulty structure is a time-varying function. According to J.C. theory, the reliability index within any time period t is, in fact, a standard normal distribution in a Cartesian coordinate system. The perpendicular distance between the origin and the line representing the limit state equation X = S within the normally distributed rectangular coordinate system XOS is, for example... Figure 4 As shown.

[0145] The algorithm steps for calculating the reliability index are as follows:

[0146] (a) Calculate the state output vector X of any measured device within time period t. i and its corresponding evaluation norm vector S i mean And set the initial coordinates of its limit verification point as follows:

[0147] (b) Test each X individually i S i If the distribution conforms to a normal distribution, proceed directly to step (c); otherwise, check the calculation point P according to formulas (7) to (11). * [X * ,S *Perform equivalent normalization at the [] position to obtain the values ​​of each vector. Then proceed to the next step.

[0148] (c) Calculate the direction cosine according to formulas (4) and (5): and

[0149] (d) Calculate using formula (3) (i = 1, 2, ..., n).

[0150] (e) Substitute the above results into the corresponding limit state equation to obtain the structural reliability index β.

[0151] (f) Repeat steps (c) to (e) and compare the obtained reliability index β value with the previous round's calculated value until the difference between the values ​​of β in two adjacent rounds is less than the set allowable value (the allowable value is set according to the actual situation and is not specifically limited here). In this way, the reliability index β for this period is obtained.

[0152] (g) By analogy, the structural reliability index β(t) for each analysis period within the preset lifespan T can be obtained. j )sequence:

[0153] {β(t j )},t j ≤T,j=1,2,3,…,n (12).

[0154] When the device status output vector X = [X1, X2, ..., X...] n S = [S1, S2, ..., S] n When a function conforms to a normal distribution, its function Z = g(X,S) also conforms to a normal distribution. For example... Figure 5 The failure probability of the normal distribution function Z and its reliability β have a precise mathematical relationship as shown in formulas (13) and (14):

[0155]

[0156] Among them, P f σ represents the failure probability. Z μ Z Let Z represent the variance and mean of the safety assessment limit corresponding to the normal distribution function Z, respectively, and Φ(-β) represent the failure probability as a function of the reliability index.

[0157] As shown in formula (14), there is a one-to-one correspondence between the failure probability of the equipment and the reliability index β. By obtaining the reliability index β, the failure probability can be obtained, and thus the equipment reliability probability can be obtained:

[0158] Ps =1-P f =1-Φ(-β) (15);

[0159] Among them, P s This indicates the probability of equipment reliability.

[0160] Step S304: Establish the equipment remaining life distribution function based on the failure probability, and perform curve fitting on the equipment remaining life distribution function to obtain the equipment remaining service life.

[0161] Specifically, step S304 includes:

[0162] Step S3041: Establish a failure distribution function based on the failure probability, and transform the failure distribution function into a remaining life distribution function of the equipment based on the conditional probability formula.

[0163] Specifically, the remaining lifespan T of a device that has been operating safely for t hours under fault-free conditions, and can continue to operate until failure or malfunction occurs, is called the remaining lifespan T of the device with a service life of t. t Therefore, the remaining lifetime T t It is a random variable. It is a conditional probability, and the distribution function of the remaining equipment life is expressed as:

[0164] F t (x)=P(T t ≤x)=P(T≤t+x|T>t)

[0165]

[0166] Among them, F t (x) represents the equipment's remaining lifetime distribution function, and x represents the equipment's state parameters.

[0167] If we assume its failure distribution function is R(t), then the above equation can be transformed into:

[0168]

[0169] Step S3042: Use a curve fitting tool to fit the remaining life distribution function of the equipment, and calculate the remaining life of the equipment based on the curve fitting.

[0170] Specifically, based on the distribution function of the remaining useful life of the equipment, the distribution law of the known remaining useful life of the equipment is obtained. A fitting curve is then generated using fitting tools such as MATLAB to calculate the remaining useful life of the equipment. The specific steps for generating the fitting curve using MATLAB or similar tools can be found in relevant technical documents and will not be elaborated upon here.

[0171] The bearing reliability calculation and remaining life prediction method provided in this embodiment derives the remaining life distribution function based on the failure probability and conditional probability formulas, strictly adhering to probability and statistics theory to ensure that the distribution function accurately characterizes the random characteristics of the remaining life, providing a reliable mathematical model for subsequent predictions. By fitting the distribution function with a curve fitting tool, the abstract probability distribution is transformed into a directly readable remaining life value, achieving a leap from qualitative analysis to quantitative prediction. This provides an accurate time reference for power grid task planning that requires continued operation after unit failure, improving the scientific nature of operation and maintenance decisions. Directly linking the failure probability with the remaining life distribution function creates a closed loop between reliability index calculation, failure probability analysis, and remaining life prediction, ensuring logical consistency between the life prediction results and previous reliability assessments, thus enhancing the systematicness and coherence of the overall method.

[0172] As one or more specific application embodiments of the present invention, combined with Figures 6 to 17 The reliability calculation and remaining life prediction of the bearing bush provided by the present invention are further described in detail below:

[0173] Unit #2 of a power plant is a subcritical steam turbine generator set of type N335-16.18 / 538 / 538. The high, medium, and low-pressure rotors are connected by rigid couplings. The generator-exciter couplings are semi-flexible wave couplings, the exciter-exciter couplings are flexible couplings, and the exciter bearings are spherical bearing pads with slotted upper bearings, B / D = 0.32 (range [0.3, 2]), and ψ = c / R = 0.004 elliptical bearings. The shaft system structure is as follows: Figure 6 As shown.

[0174] At approximately 7:00 AM on March 23, 2023, an abnormal noise suddenly occurred near the generator's auxiliary exciter. At this time, the unit load was 285MW. The vibration levels of bearings #11 on both sides of the auxiliary exciter were as follows: vertical vibration 52μm (according to on-site practice, the measured acceleration values ​​are converted to displacement values, the same applies below), horizontal vibration 71μm, and axial vibration 38μm; bearing #12 showed vertical vibration 25μm, horizontal vibration 54μm, and axial vibration 34μm. Investigation revealed that the vibration levels of bearings #11 and #12 had been consistently high (with horizontal vibration around 60μm (standard ≤50μm)) for nearly two months prior to the abnormal noise, but were relatively stable. Furthermore, a shaft oil temperature test was conducted after the fault, and no significant changes in vibration or abnormal noise were observed. The unit's shaft system has an online vibration monitoring system installed on bearings #1 to #10, but this device is not present on bearings #11 and #12 on both sides of the auxiliary exciter. During the failure, handheld thermometers and offline vibration analyzers were used to track and measure the bearings of #11 and #12. Since the reliability evaluation and life prediction methods for vibration and temperature are the same, for the sake of convenience, only vibration will be used as an example for explanation below, and the bearing temperature evaluation will be omitted.

[0175] Analysis of on-site measurement data in the early stage of the fault revealed the following vibration characteristics: (1) The amplitude of bearings #11 and #12 exceeded the national standard value; (2) The frequency spectrum of bearings #11 and #12 showed obvious high-frequency components such as 2nd and 3rd harmonics, and the amplitude in some directions was much greater than the 1st harmonic component; among them, the main vibration frequencies of bearing #11 were the 1st harmonic and 0.5th harmonic; bearing #12 was mainly the 1st harmonic accompanied by 0.48th and 0.51st harmonics, as shown in the figure below. Figure 7 and Figure 8 As shown. (3) The time-domain waveforms measured at both bearings exhibit irregular periodic impacts, with 3 to 5 time-domain waveforms per acquisition cycle (including the time-domain waveforms of the low-frequency and high-frequency ranges of bearing #11 and bearing #12), and the specific waveforms are as follows. Figures 9 to 12 As shown.

[0176] Depend on Figure 13 and Figure 14 As shown, the time-frequency characteristics of two adjacent sampling periods change with time through a "translation," but both the amplitude and phase of the vibration change significantly during the translation process, and the number of waveforms per unit revolution also shows significant differences; that is, whether a certain peak appears in each revolution and its amplitude change are highly random. In addition, as the subsequent measurement process continues, the time interval between the waveforms of two adjacent periods also changes significantly, and the amplitude value occasionally exhibits a "step" phenomenon between adjacent intervals.

[0177] In summary, the vibration time-domain waveforms of bearings #11 and #12 exhibit significant nonlinear random characteristics, both in their cyclic history per unit revolution and in their time history as a function of state changes. Their vibration spectra also display broad-spectrum characteristics characteristic of loosening, impact, rubbing, and swaying. Therefore, bearings #11 and #12 can be diagnosed as fracture and fragmentation faults. Figure 15 The figure shown is a diagram illustrating the measured vibration of bearing #11 in various directions.

[0178] Based on the degree of correlation with bearing breakage, the vertical, horizontal, and axial vibrations of the faulty bearing and its adjacent bearings, as well as the bearing cap temperature of the faulty bearing (no bearing temperature and return oil temperature measuring devices were available on-site), were determined as the critical path output vectors. First, frequent, timed sampling was required, along with simultaneous reliability tracking calculations. After accumulating a certain amount of data, the sampling period was determined based on the early characteristics of the vectors. Once the data accumulation reached a point where a relatively stable reliability fitting curve could be formed, timed sampling and tracking were then conducted in segments according to the predicted remaining safe life, until a critical output vector approached the failure probability trigger condition. To ensure equipment and personnel safety, the initial sampling interval was set at 15 minutes, with each sample acquisition time approximately 3 seconds (determined by the limitations of the on-site sampling equipment).

[0179] According to relevant technologies, the main factors causing damage to sliding bearing bushes due to abnormal vibrations in the turbine generator set shaft system include: unreasonable load distribution, rotor center misalignment, journal-bearing rubbing or incorrect dynamic-static center alignment, poor spherical contact between the bearing pads and their recesses, deformation of the corrugated flexible coupling, poor tungsten alloy material or casting quality, insufficient bearing housing series stiffness, abnormal rotor vibration, or changes in input load. Looking at Unit #2, apart from vibration and noise abnormalities in bearings #11 and #12, the vibration and bearing temperature at other bearing locations in the shaft system did not fluctuate significantly, indicating that the unit fault occurred locally and could only have a local impact. Therefore, the preliminary screening results for the faulty equipment status output vector are as follows:

[0180]

[0181] The corresponding evaluation norm vector is as follows:

[0182]

[0183] In the formula, This indicates the vibration in the d direction of bearing #k or the metal temperature of that bearing. This represents the expert-judged limit value for vibration in the d direction of bearing #k or the metal temperature of that bearing, where k = 11, 12, and d = ⊥, -, ⊕ (representing vertical, horizontal, and axial directions, respectively).

[0184] Since the horizontal vibration of turbines #11 and #12 exceeds the national standard's acceptable value, a safety evaluation specification for the corresponding output vector must be developed independently of the faulty equipment, based on the operating experience of similar units, the manufacturer's design margin, and the experience of on-site experts. Clearly, the vector and the specification are independent of each other and satisfy a normal distribution, thus satisfying the limit state equation: Z(t)=g[X(t),S(t)]=0.

[0185] Furthermore, as can be seen from the aforementioned fault description, bearings #11 and #12 failed one after the other, indicating a correlation and causal effect between them. Additionally, correlation analysis of the vibrations in each direction of the same bearing bush based on the output vector data measured at the initial stage of the fault revealed that the correlation factors between each direction were close to 1, indicating that the vibration levels in each direction were closely related and not independent. Therefore, the equipment status output vector in formula (18) can be simplified to the vibration level in a typical direction and the metal temperature of the bearing bush, forming the t-th vibration vector of the bearing bush system in this fault. j (j=1,2,…,m,…) State evaluation space for the detection time period:

[0186]

[0187] In the formula:

[0188]

[0189] in, This represents the vector of horizontal vibration amplitude values ​​for tile #11. This indicates the metal temperature of the bearing cap shell of bearing #11. This represents the limiting vector of the horizontal vibration amplitude value of the #11 watt. This represents the temperature limit vector of the outer metal of the tile cover of #11.

[0190] Therefore, the limit state equations for vibration and bearing temperature are as follows:

[0191]

[0192] Substituting into equations (1) to (6) and following the iterative steps of the JC method, t is obtained respectively. j During the time period Then, the failure probability at each time period is obtained. The design specifications for equipment condition evaluation are as follows:

[0193]

[0194] In the formula, Indicates up to t j Before the analysis period (including t) j The overall failure probability for all time periods. Indicates by t j Previous analysis periods (including t) j (analysis period) The reliability of the horizontal vibration vector obtained by fitting at the current evaluation time. Indicates by t j Each preceding time period (including t) j (analysis period) The reliability of the fitted temperature vector of the tile shell metal at the current evaluation moment. Indicates the analysis period t j The sum of all previous analysis periods.

[0195] Based on the discrete analysis principle of time-varying bearing reliability and formula (26), the comprehensive evaluation logic of the bearing system at the current moment is designed as follows:

[0196]

[0197] In the formula, β0: the allowable limit value of equipment state reliability given by experts. If β0 is the limit reliability index given by experts based on experience that makes the probability of equipment failure 1, that is, when P fWhen Φ(-β0) = 1, the equipment fails. According to formula (27), the failure time T0 can be calculated, and the equipment's ultimate remaining life can be obtained from this.

[0198] Δt=T0-Γ j (28).

[0199] If β0 is not the limiting reliability index that makes the probability of equipment failure 1, then the measured data should be used to obtain... This will generate a fitted curve: Find the time (T′0) corresponding to the intersection of the two curves with the β0 line. — (T′0) T Then the remaining safe lifetime of the two output vectors is:

[0200]

[0201] Take the smaller of the two as the remaining safe life of the equipment:

[0202] Δt=min[(Δt) — ,(Δt) T (30).

[0203] Since the #11 and #12 watts of this unit lack online measurement devices and cannot continuously obtain output vector measurement information, a method of using typical samples to represent time period information is adopted. That is, during a certain time period or a relatively stable equipment state stage, offline measurement devices are used to continuously sample until the instrument's specified maximum single-sample data length limit (approximately 3 seconds in this example). The statistical regularity of this sample is then used to represent the statistical regularity of that time period for calculation and analysis. Furthermore, on-site experts set corresponding constraint standards for the output vector based on the equipment's real-time state. For example, in this example, relative to the amplitude of the bearing's horizontal vibration, its upper constraint limit for that time period can be set as: [(mean value of measured horizontal amplitude samples for that time period / mean value of vertical amplitude for that time period)] × mean value of expert-experienced limit values ​​for vertical amplitude, i.e., (Note: Expert experience limits refer to random quantities that follow a normal distribution and generally do not exceed 110% of the acceptable value.) A flowchart illustrating the specific principle of the sample time period analysis method is shown below. Figure 16 As shown.

[0204] Offline instrument field testing began at 7:15 AM on March 23, 2023. Initially, testing was conducted in 15-minute intervals. Sampling was performed at corresponding times within each interval (generally near the end of the interval), and the reliability of that interval, along with the reliability of all intervals up to that point, was calculated. This process was repeated. Based on the measured output vectors (horizontal amplitude, tile shell temperature) of the #11 tile at each interval and the expert-given evaluation limits, the reliability of each interval was calculated, and a reliability trend curve was fitted. Finally, combined with the expert-given reliability tolerance limits, the calculation formula is as follows:

[0205]

[0206] In the formula, This indicates the maximum permissible vertical amplitude limit (defined value) of the bearing bush. Indicates t j The mean of all measured vertical amplitudes before the analysis period is used. Indicates t j The maximum value of all horizontal amplitudes during and before the analysis period. This indicates the acceptable value specified in the standard for vertical vibration of the bearing housing. t j The average horizontal amplitude of the samples during the analysis period Indicates t j The average vertical amplitude of the samples during the analysis period.

[0207] Based on the above calculations, the safe remaining service life of the sliding bearing is finally predicted, and the results are as follows: Figure 17 As shown. By Figure 17 It is known that the estimated remaining life end point of the bearing is controlled by the bearing temperature reliability, while the actual downtime exceeds this control point, ensuring both grid demand and unit safety. The remaining life assessment method has also been practically verified. However, due to grid safety and construction schedule constraints, limit control specifications and minimum reliability tolerances were imposed on-site, resulting in severe damage to the bearing.

[0208] The bearing reliability calculation and remaining life prediction method provided in this embodiment determines the occurrence of bearing breakage failure based on the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics of the bearing vibration time-domain waveform. Considering that the unit must continue to operate for a period of time to complete specific tasks of the power grid, a time-varying reliability bearing remaining life prediction algorithm based on the JC method (verification point method) is designed, using the information state equation as the theoretical basis. First, a random response limit state equation based on arbitrary distribution is established. Then, the reliability and comprehensive failure probability of the fault completeness vector are calculated in different time periods. Finally, the remaining life prediction result is obtained through curve fitting, achieving the goal of more convenient, faster, and more effective analysis and calculation of equipment reliability and remaining life.

[0209] This embodiment also provides a bearing reliability calculation and remaining life prediction device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described previously. As used below, the term "module" can be a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0210] This embodiment provides a device for calculating bearing reliability and predicting remaining life, such as... Figure 18 As shown, it includes:

[0211] The vector acquisition module 1801 is used to acquire the equipment status output vector and the corresponding evaluation standard vector of the faulty equipment when the bearing bush breaks.

[0212] The limit state equation establishment module 1802 is used to determine whether the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, and to directly establish the limit state equation based on the judgment result, or to establish the limit state equation based on the preset limit verification point.

[0213] The reliability and failure probability calculation module 1803 is used to calculate the reliability index of the equipment's preset life cycle based on the limit state equation, and to calculate the corresponding failure probability based on the reliability index.

[0214] The equipment remaining useful life prediction module 1804 is used to establish the equipment remaining useful life distribution function based on the failure probability, and to perform curve fitting on the equipment remaining useful life distribution function to obtain the equipment remaining useful life.

[0215] In some optional implementations, the bearing reliability calculation and remaining life prediction device further includes:

[0216] The bearing breakage fault judgment module is used to acquire the time-domain waveform of the equipment bearing vibration and extract the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics of the bearing vibration time-domain waveform; when the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics are distorted, the bearing breakage fault is determined.

[0217] In some alternative implementations, the vector acquisition module 1801 includes:

[0218] The vector acquisition unit is used to extract the output vector corresponding to the specified state of the faulty equipment based on the preset information state equation, so as to obtain the equipment state output vector and the corresponding evaluation standard vector. The evaluation standard is a comprehensive evaluation space formed based on relevant standards, design data and operating experience of similar equipment or relevant expert experience.

[0219] In some optional implementations, the limit state equation establishment module 1802 includes:

[0220] The first establishment unit is used to directly establish the limit state equation based on the device state output vector and the corresponding evaluation standard vector when the device state output vector and the corresponding evaluation standard vector conform to a normal distribution.

[0221] The second establishment unit is used to perform equivalent normalization processing on the device state output vector and the corresponding evaluation standard vector based on preset limit verification points when the device state output vector and the corresponding evaluation standard vector do not conform to the normal distribution, to obtain a normally distributed random vector, and to establish the limit state equation based on the normally distributed random vector.

[0222] In some alternative implementations, the second establishing unit includes:

[0223] The equivalent normalization subunit is used to calculate the distribution function and probability density of the non-normally distributed device state output vector and the corresponding evaluation standard vector at the preset limit verification point. Based on the corresponding distribution function and probability density, the equivalent normalization transformation is performed using the standard normal distribution function to obtain a normally distributed random vector.

[0224] In some optional implementations, the reliability and failure probability calculation module 1803 includes:

[0225] The time period division unit is used to divide the preset life cycle of the equipment into multiple analysis time periods.

[0226] The first reliability index calculation unit is used to perform a standard transformation on the limit state equation of each analysis period when the equipment state output vector and the corresponding evaluation standard vector of each analysis period conform to a normal distribution, so as to obtain the standard limit state surface of each analysis period. The shortest distance from the origin of the preset dimension space standard normal coordinate system to the standard limit state surface is used as the reliability index within the preset life cycle of the equipment.

[0227] The second reliability index calculation unit is used to initialize the preset limit verification point coordinates when the equipment status output vector and the corresponding evaluation standard vector of each analysis period do not conform to the normal distribution, calculate the direction cosine sum based on the preset limit verification point coordinates, update the limit verification point coordinates using the direction cosine sum; substitute the updated limit verification point coordinates into the limit state equation of each analysis period to iteratively calculate the equipment reliability, and combine the reliability of all analysis periods to obtain the reliability index.

[0228] In some alternative implementations, the device remaining useful life prediction module 1804 includes:

[0229] The distribution function establishment unit is used to establish a failure distribution function based on the failure probability and to transform the failure distribution function into a remaining life distribution function of the equipment based on the conditional probability formula.

[0230] The fitting unit is used to fit the remaining life distribution function of the equipment using a curve fitting tool, and to calculate the remaining life of the equipment based on the curve fitting.

[0231] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0232] In this embodiment, the bearing reliability calculation and remaining life prediction device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0233] This invention also provides a computer device having the above-described features. Figure 18 The device shown is for calculating bearing reliability and predicting remaining life.

[0234] Please see Figure 19 , Figure 19 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 19 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 19 Take a processor 10 as an example.

[0235] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0236] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0237] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0238] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0239] The computer device also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 can be connected via a bus or other means. Figure 19 Taking the example of a connection between China and Israel via a bus.

[0240] Input device 30 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the computer device, such as a touchscreen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 40 may include display devices, auxiliary lighting devices (e.g., LEDs), and haptic feedback devices (e.g., vibration motors). The aforementioned display devices include, but are not limited to, liquid crystal displays, light-emitting diodes, displays, and plasma displays. In some alternative embodiments, the display device may be a touchscreen.

[0241] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0242] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0243] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for calculating bearing reliability and predicting remaining life, characterized in that, The method includes: When a bearing bush breaks, the equipment status output vector of the faulty equipment and the corresponding evaluation standard vector are obtained. Determine whether the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, and directly establish the limit state equation based on the judgment result, or establish the limit state equation based on the preset limit verification point. The reliability index of the equipment for the preset life cycle is calculated based on the limit state equation, and the corresponding failure probability is calculated based on the reliability index. Based on the failure probability, a remaining life distribution function of the equipment is established, and the remaining life distribution function of the equipment is curve-fitted to obtain the remaining life of the equipment.

2. The method according to claim 1, characterized in that, The method further includes determining whether a bearing bush has fractured by means of the following: The time-domain waveform of the equipment bearing vibration is obtained, and the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics of the bearing vibration time-domain waveform are extracted. When the nonlinear random characteristics and wide-spectrum amplitude-frequency characteristics are distorted, it is determined that the bearing has broken.

3. The method according to claim 1, characterized in that, Obtain the device status output vector and corresponding evaluation standard vector of the faulty device, including: Based on the preset information state equation, the output vector corresponding to the specified state of the faulty equipment is extracted to obtain the equipment state output vector and the corresponding evaluation standard vector; the evaluation standard is an evaluation space formed by comprehensively considering relevant standards, design data, and operating experience of similar equipment or relevant expert experience.

4. The method according to claim 1, characterized in that, The method of directly establishing the limit state equation based on the judgment result, or establishing the limit state equation based on a preset limit verification point, includes: When the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, the limit state equation is directly established based on the device state output vector and the corresponding evaluation standard vector. When the device state output vector and the corresponding evaluation standard vector do not conform to the normal distribution, the device state output vector and the corresponding evaluation standard vector are subjected to equivalent normal processing based on the preset limit verification point to obtain a normally distributed random vector, and the limit state equation is established based on the normally distributed random vector.

5. The method according to claim 4, characterized in that, The process of performing equivalent normalization on the device state output vector and the corresponding evaluation standard vector based on preset limit verification points to obtain a normally distributed random vector includes: At the preset limit verification point, calculate the distribution function and probability density of the non-normally distributed device state output vector and the corresponding evaluation standard vector. Based on the corresponding distribution function and probability density, the standard normal distribution function is used to perform equivalent normalization transformation to obtain a normally distributed random vector.

6. The method according to claim 4, characterized in that, The calculation of the reliability index of the equipment within a preset lifespan based on the limit state equation includes: The equipment's preset lifespan is divided into multiple analysis periods; When the device status output vector and the corresponding evaluation standard vector for each analysis period conform to a normal distribution, a standard transformation is performed on the limit state equation for each analysis period to obtain the standard limit state surface for each analysis period. The shortest distance from the origin of the preset dimension space standard normal coordinate system to the standard limit state surface is used as the reliability index within the preset life cycle of the device. When the device status output vector and the corresponding evaluation standard vector for each analysis period do not conform to a normal distribution, the coordinates of the preset limit verification point are initialized, the direction cosine sum is calculated based on the preset limit verification point coordinates, and the limit verification point coordinates are updated using the direction cosine sum; the updated limit verification point coordinates are substituted into the limit state equation for each analysis period to iteratively calculate the device reliability, and the reliability of all analysis periods is combined to obtain the reliability index.

7. The method according to claim 1, characterized in that, Based on the failure probability, a remaining lifespan distribution function is established, and curve fitting is performed on the remaining lifespan distribution function to obtain the remaining lifespan of the equipment, including: A failure distribution function is established based on the failure probability, and the failure distribution function is transformed into a remaining life distribution function of the equipment based on the conditional probability formula; The remaining lifespan distribution function of the equipment is fitted using a curve fitting tool, and the remaining lifespan of the equipment is calculated based on the curve fitting.

8. A device for calculating bearing reliability and predicting remaining life, characterized in that, The device includes: The vector acquisition module is used to acquire the equipment status output vector and the corresponding evaluation standard vector of the faulty equipment when the bearing bush breaks. The limit state equation establishment module is used to determine whether the device state output vector and the corresponding evaluation standard vector conform to a normal distribution, and to directly establish the limit state equation based on the judgment result, or to establish the limit state equation based on a preset limit verification point. The reliability and failure probability calculation module is used to calculate the reliability index of the equipment for the preset life cycle based on the limit state equation, and to calculate the corresponding failure probability based on the reliability index. The equipment remaining service life prediction module is used to establish a distribution function of equipment remaining service life based on the failure probability, and to perform curve fitting on the distribution function of equipment remaining service life to obtain the equipment remaining service life.

9. A computer device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the bearing reliability calculation and remaining life prediction method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the bearing reliability calculation and remaining life prediction method according to any one of claims 1 to 7.

Images