A method for estimating equipment remaining life by integrating grey prediction modeling and discrete random impact damage

Abstract

The present invention discloses a method for estimating the remaining life of equipment that integrates grey prediction modeling and discrete random impact damage, including: collecting degradation data of equipment performance indicators, analyzing the data to extract degradation data characteristics; based on the degradation data characteristics, finding the equipment performance degradation characteristics, building a grey prediction model that integrates random impacts, and estimating the remaining life of the equipment; solving the grey prediction model parameters, and solving the impact characteristics including the first impact number and the average degradation trajectory, and then calculating the probability of equipment failure; based on the impact characteristics, solving the probability distribution of the remaining life of the equipment, calculating the equipment reliability, and estimating the average remaining life of the equipment. The present invention effectively solves the problem of equipment remaining life estimation and health management in small sample discrete impact scenarios, can accurately characterize the performance degradation law of equipment under discrete random impacts, significantly improve the accuracy of remaining life estimation, and provide a reliable decision-making basis for scientific equipment management.

Description

Technical Field

[0001] The present invention belongs to the technical field of equipment health management, and in particular relates to a method for estimating the remaining life of equipment that integrates grey prediction modeling and discrete random impact damage. Background Art

[0002] Large, complex equipment often faces a combination of random shocks from multiple sources, such as vibration, stress, and temperature, during service. These external stimuli can continuously degrade equipment performance. The timing and intensity of these random shocks exhibit significant uncertainty over time. However, due to the highly integrated nature of equipment systems and the complexity of their internal structures, real-time monitoring of performance degradation of core components is difficult to obtain, rendering existing remaining life prediction methods inapplicable. However, developing a scientific remaining life prediction model and accurately analyzing equipment life distribution and reliability is not only a key foundation for condition-based maintenance and dynamic management of equipment, but also a core component in improving reliability and performance assessment throughout the equipment's lifecycle. This is crucial for ensuring the safe operation of equipment and the completion of production tasks.

[0003] Random shocks to equipment during operation are the result of a complex interplay of uncertainties from multiple sources. This uncertainty encompasses both the randomness of the shock's moment of occurrence and the volatility of the resulting performance degradation. Each shock can potentially cause equipment performance degradation, and as the number of shocks increases, it can lead to equipment failure. Notably, the uncertainty surrounding the shock's arrival exhibits typical random characteristics, and the number of shocks exhibits the characteristics of "small sample size and poor information" (some equipment experiences only a small number of shocks, such as ships impacted by extreme waves). Traditional probabilistic and statistical methods often face the challenge of insufficient data when addressing such issues. However, the gray prediction model, with its powerful ability to process limited data, can effectively extract the degradation trends implicit in discrete shock sequences, providing a scientifically descriptive tool for quantifying equipment performance degradation. Compared to traditional probabilistic and statistical methods, the gray prediction model demonstrates significant advantages under these "small sample size and poor information" conditions. Traditional methods rely on extensive historical data to establish probability distribution models. However, in real-world projects, due to the short service life and low frequency of extreme operating conditions, many equipment models struggle to accumulate sufficient data samples, resulting in significant bias in prediction results. The grey prediction model transforms discrete shock data into a regular sequence, enabling dynamic tracking and prediction of equipment degradation. Combining deterministic trend prediction with random fluctuation analysis can comprehensively characterize equipment degradation patterns under random shocks.

[0004] To accurately characterize this complex uncertainty, this paper combines stochastic processes with grey prediction models to form a new hybrid modeling framework. First, the original discrete device performance degradation data is accumulated to reduce randomness and uncover potential trends in device performance degradation. Simultaneously, based on stochastic process theory, a Poisson process or update process is used to probabilistically model the impact arrival interval, quantifying the randomness of the impact's occurrence. This organic fusion of the two allows the model to capture both the deterministic trends in device performance degradation and the stochastic nature of the impact process, enabling dynamic and accurate prediction of the device's remaining lifespan.

[0005] Based on the above modeling ideas, combined with the Poisson process and grey prediction models, a quantitative relationship is established between the equipment performance degradation index and the remaining life, the equipment performance degradation trend is predicted, and the failure probability of the equipment at different times is calculated using the theory, thereby achieving dynamic prediction of the equipment's remaining life and providing a reliable decision-making basis for scientific equipment management. Summary of the Invention

[0006] To address the challenges of existing technologies, this paper proposes a method for estimating the remaining life of equipment using gray prediction modeling and discrete random impact damage. This method addresses the problem of estimating the remaining life of equipment in scenarios with a small number of discrete random impacts. It constructs a discrete gray prediction model incorporating a Poisson process. It solves the expression for the first number of impacts that affect equipment performance degradation, the average degradation trajectory, and calculates the probability of equipment failure to obtain the probability distribution of the equipment's remaining life, ultimately forming a method for estimating the remaining life of equipment in the presence of discrete random impact damage. Compared to traditional methods, the gray prediction model overcomes the limitations of data dependence and can effectively model even in the extreme case of a small number of samples. This presents a novel prediction approach for complex equipment with scarce life data and high monitoring costs. Traditional statistical modeling methods (such as the Weibull distribution) suffer from parameter estimation bias due to data sparsity. Physical-based life prediction requires an accurate characterization of processes such as material aging and thermal stress fatigue in satellite components, making it difficult to implement in practice. Compared with the existing technology, the present invention aims to solve the problem of equipment remaining life estimation and health management in small sample discrete shock scenarios, so as to accurately characterize the performance degradation law of equipment under discrete random shocks, significantly improve the accuracy of remaining life estimation, and provide a reliable decision-making basis for scientific equipment management.

[0007] In order to achieve the above technical objectives, the present invention provides the following technical solutions:

[0008] A method for estimating the remaining life of equipment by integrating grey prediction modeling and discrete random impact damage specifically comprises the following steps:

[0009] S1. Collect degradation data of equipment performance indicators, analyze the degradation data and extract the characteristics of the degradation data;

[0010] S2. Based on the characteristics of degradation data, find the characteristics of equipment performance degradation and build a grey prediction model that integrates random shocks to estimate the remaining life of the equipment;

[0011] S3. Solve the parameters of the grey prediction model constructed in step S2, and solve the impact characteristics based on the parameters of the grey prediction model, including: the number of first impacts and the average degradation trajectory; then calculate the equipment failure probability;

[0012] S4. Based on the impact characteristics, solve the probability distribution of the remaining life of the equipment, calculate the equipment reliability, and estimate the average remaining life of the equipment.

[0013] Furthermore, step S1 is specifically as follows:

[0014] The degradation data of equipment performance indicators are collected and sorted, and discrete time series are used to represent the performance degradation of the equipment after being subjected to discrete random shocks. Each discrete random shock is recorded as a moment; from the first moment to the n The equipment performance index degradation sequence at time is recorded as: ;

[0015] At the same time, the cumulative degradation of the equipment performance indicators after each impact is calculated, and the formula is expressed as:

[0016] ;

[0017] in, Indicates the degradation of the equipment performance index at time i.

[0018] Furthermore, step S2 is specifically as follows:

[0019] The cumulative degradation of the equipment performance index obtained in step S1 , analyze the growth trend of performance index degradation after discrete random shock, and construct a discrete random grey prediction model. The formula is expressed as:

[0020] ;

[0021] in, For equipment in time period The number of impacts received, The impact strength parameter is Homogeneous Poisson process; 、 are the two parameters to be solved.

[0022] Furthermore, step S3 specifically includes:

[0023] S31: The cumulative degradation of the equipment performance index obtained in step S1 , solve the parameters of the random discrete grey prediction model, and its calculation formula is as follows:

[0024] ;

[0025] in, ; ;

[0026] S32. Based on the discrete random grey prediction model and the failure threshold of the equipment performance index, the first impact number T is solved. The specific calculation formula is:

[0027] ;

[0028] in, is the critical value of equipment performance indicator failure, is the initial value of the equipment performance index degradation, is the value function;

[0029] S33. Based on the number of first impacts, the discrete random grey prediction model and the properties of the Poisson process, the average degradation trajectory of the equipment performance degradation index is solved. It is expressed as the mathematical expectation of the degradation amount, and the formula is expressed as follows:

[0030] ;

[0031] S34. According to the number of first-arrival shocks and the properties of the Poisson process, we can get affected The probability of a random shock is:

[0032] ;

[0033] In the time period The probability of equipment failure is:

[0034] ;

[0035] in, is the impact strength parameter.

[0036] Furthermore, step S4 specifically includes:

[0037] S41. Calculate the distribution of the remaining life of the equipment based on the random discrete grey prediction model that follows the equipment performance degradation law. Gamma distribution , its probability density function is:

[0038] ;

[0039] Its reliability function is:

[0040] ;

[0041] S42. Distribution based on the remaining life of the equipment , calculate the mathematical expectation of the distribution:

[0042] ;

[0043] S43, according to the impact arrival interval time , estimate the impact strength parameters:

[0044] ;

[0045] For the device to be The first shock and the The time interval between shocks, The total number of times the device is impacted;

[0046] S44. Substitute the estimated value of the impact intensity parameter calculated in step S43 and the number of first impacts calculated in step S32 into the expected distribution of the remaining life of the equipment to obtain the estimated value of the average remaining life of the equipment.

[0047] The present invention further provides a device, comprising a memory and a processor, wherein:

[0048] a memory for storing computer programs capable of running on the processor;

[0049] The processor is used to execute the above-mentioned method for estimating the remaining life of equipment by integrating grey prediction modeling and discrete random impact damage when running the computer program.

[0050] In addition, according to the above-mentioned device, the present invention also provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the processor to implement the above-mentioned method for estimating the remaining life of equipment that integrates gray prediction modeling and discrete random impact damage when executed.

[0051] Based on the above technical solution, the present invention has at least the following beneficial effects:

[0052] This paper addresses the randomness and non-repeatability of impact events, the nonlinear characteristics of equipment performance degradation, and the multi-factor coupling effects of discrete random shocks. It proposes a method for estimating equipment's remaining life that integrates grey prediction modeling with discrete random shock damage. By quantifying the occurrence patterns of random shocks and the trends in equipment performance degradation, a model for predicting equipment performance degradation and remaining life suitable for discrete time shock scenarios is constructed. This method derives the number of impacts that first cause equipment performance degradation and its failure probability, providing a probability distribution of the equipment's remaining life and enabling equipment remaining life prediction. This effectively addresses the problem of equipment remaining life estimation and health management in small-sample discrete shock scenarios, accurately characterizing equipment performance degradation patterns under discrete random shocks. This significantly improves the accuracy of equipment remaining life estimation, helps equipment managers better understand equipment operation, and provides a reliable basis for decision-making in scientific equipment management. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 Schematic diagram of the process of the method proposed in the present invention;

[0054] Figure 2 Calculate the equipment performance degradation prediction curve and the first impact times;

[0055] Figure 3 The equipment failure probability curve at different impact intensity levels;

[0056] Figure 4 The probability density function diagram of the remaining life distribution of equipment under different impact intensity levels. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the present invention more clear, the following Figure 1-4 It should be understood that the specific embodiments described herein are only used to illustrate the present invention and are not intended to limit the present invention.

[0058] Although the steps in the present invention are arranged with numbers, they are not intended to limit the order of the steps. Unless the order of the steps is clearly stated or the execution of a step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" used herein refers to and covers any and all possible combinations of one or more of the associated listed items.

[0059] like Figure 1 As shown in the figure, the present invention proposes a method for estimating the remaining life of equipment that integrates grey prediction modeling and discrete random impact damage. The method includes two parts: drawing a prediction curve for equipment performance degradation and calculating the remaining life distribution. Specifically, it includes the following steps:

[0060] S1. Collect degradation data of equipment performance indicators, analyze the degradation data and extract the characteristics of the degradation data;

[0061] As a preferred embodiment, step S1 is specifically as follows:

[0062] The degradation data of equipment performance indicators are collected and sorted, and discrete time series are used to represent the performance degradation of the equipment after being subjected to discrete random shocks. Each discrete random shock is recorded as a moment; from the first moment to the n The equipment performance index degradation sequence at time is recorded as: ;

[0063] At the same time, the cumulative degradation of the equipment performance indicators after each impact is calculated, and the formula is expressed as:

[0064] ;

[0065] in, Indicates the degradation of the equipment performance index at time i.

[0066] S2. Based on the characteristics of degradation data, find the characteristics of equipment performance degradation and build a grey prediction model that integrates random shocks to estimate the remaining life of the equipment;

[0067] As a preferred embodiment, step S2 is specifically as follows:

[0068] The cumulative degradation of the equipment performance index obtained in step S1 , analyze the growth trend of performance index degradation after discrete random shock, and construct a discrete random grey prediction model. The formula is expressed as:

[0069] ;

[0070] in, For equipment in time period The number of impacts received, The impact strength parameter is Homogeneous Poisson process; 、 are the two parameters to be solved.

[0071] The present invention combines random processes with grey prediction models to form a new hybrid modeling approach. First, the original discrete equipment performance degradation data is accumulated and generated to weaken the randomness of the data and explore the potential trend of equipment performance degradation. At the same time, based on the theory of random processes, a Poisson process or an update process is used to probabilistically model the impact arrival interval to quantify the randomness of the impact occurrence time. After organically integrating the two, the model can capture the deterministic trend of equipment performance degradation and describe the random characteristics of the impact process, thus achieving dynamic and accurate prediction of the remaining life of the equipment. Through the above steps S1-S2, the following can be obtained: Figure 2 The equipment performance degradation prediction curve shown in FIG. 1 can systematically describe the evolution law of performance degradation under different impact times through discrete time nodes.

[0072] S3. Solve the parameters of the grey prediction model constructed in step S2, and solve the impact characteristics based on the parameters of the grey prediction model, including: the number of first impacts and the average degradation trajectory; then calculate the equipment failure probability;

[0073] As a preferred embodiment, step S3 specifically includes:

[0074] S31: The cumulative degradation of the equipment performance index obtained in step S1 , solve the parameters of the random discrete grey prediction model, and its calculation formula is as follows:

[0075] ;

[0076] in, ; ;

[0077] S32. Based on the discrete random grey prediction model and the failure threshold of the equipment performance index, the first impact number T is solved. The specific calculation formula is:

[0078] ;

[0079] in, is the critical value of equipment performance indicator failure, is the initial value of the equipment performance index degradation, is a value function; in this application, the first impact number T is defined as the minimum number of impacts corresponding to equipment failure, and the remaining life distribution parameters of the equipment are calculated through the first impact number.

[0080] S33. Based on the number of first impacts, the discrete random grey prediction model and the properties of the Poisson process, the average degradation trajectory of the equipment performance degradation index is solved. It is expressed as the mathematical expectation of the degradation amount, and the formula is expressed as follows:

[0081] ;

[0082] In this embodiment, the degree of performance indicator degradation at any moment can be accurately simulated by constructing an average degradation trajectory;

[0083] S34. According to the number of first-arrival shocks and the properties of the Poisson process, we can get affected The probability of a random shock is:

[0084] ;

[0085] In the time period The probability of equipment failure is:

[0086] ;

[0087] in, is the impact strength parameter; at the same time, it can be drawn as Figure 3 As shown, the failure probability changes with time under different impact intensities.

[0088] S4. Based on the impact characteristics, solve the probability distribution of the remaining life of the equipment, calculate the equipment reliability, and estimate the average remaining life of the equipment;

[0089] As a preferred embodiment, step S4 specifically includes:

[0090] S41. Calculate the distribution of the remaining life of the equipment based on the random discrete grey prediction model that follows the equipment performance degradation law. Gamma distribution , its probability density function is:

[0091] ;

[0092] Similarly, we can draw Figure 4 As shown, the probability density function changes with time under different impact intensities;

[0093] Its reliability function is:

[0094] ;

[0095] S42. Distribution based on the remaining life of the equipment , calculate the mathematical expectation of the distribution:

[0096] ;

[0097] S43, according to the impact arrival interval time , estimate the impact strength parameters:

[0098] ;

[0099] For the device to be The first shock and the The time interval between shocks, The total number of times the device is impacted;

[0100] S44. Substitute the estimated value of the impact intensity parameter calculated in step S43 and the number of first impacts calculated in step S32 into the expected distribution of the remaining life of the equipment to obtain the estimated value of the average remaining life of the equipment.

[0101] At this point, the method proposed in the present invention addresses the problem of estimating the remaining life of equipment in scenarios with small samples of discrete random impacts, and constructs a discrete grey prediction model that integrates the Poisson process; by solving the expression of the first number of impacts of equipment performance degradation, the average degradation trajectory, and calculating the probability of equipment failure, the probability distribution of the remaining life of the equipment can be obtained, and then the remaining life of the equipment can be estimated.

[0102] This embodiment further provides a device, including a memory and a processor, wherein:

[0103] a memory for storing computer programs capable of running on the processor;

[0104] The processor is used to execute the above-mentioned method for estimating the remaining life of equipment by integrating grey prediction modeling and discrete random impact damage when running the computer program.

[0105] In addition, according to the above-mentioned device, the present invention also provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the processor to implement the above-mentioned method for estimating the remaining life of equipment that integrates gray prediction modeling and discrete random impact damage when executed.

[0106] So far, the overall process of the method proposed in the present invention has been introduced. The feasibility and effect of the method proposed in the present invention are verified by combining example simulation. In this embodiment, combined with the ultimate strength simulation data of the hull stiffener under extreme cyclic loads, the model constructed by the present invention is used to predict the ultimate strength and remaining life of the stiffener with a single rib (hereinafter referred to as the stiffener). The data in the example comes from publicly published literature. The article uses Abaqus software to simulate the cyclic load of the stiffener made of S355 steel. The pressure amplitude in the cyclic load is 5100 kN and the tension amplitude is 4950 kN. The prediction curve of the ultimate strength drop of the stiffener is as follows: Figure 2 The numerical experimental results show that the predicted curve of the ultimate strength reduction of the stiffened plate is very close to the actual value, which reflects the effectiveness and rationality of the method proposed in this invention.

[0107] In summary, the method proposed in this paper addresses the randomness and non-repeatability of impact events, the nonlinear characteristics of device performance degradation, and the multi-factor coupling effect of discrete random shocks. It proposes a device remaining life estimation method that integrates grey prediction modeling with discrete random shock damage. By quantifying the occurrence patterns of random shocks and device performance degradation trends, a device performance degradation and remaining life prediction model suitable for discrete time shock scenarios is constructed. The first impact number and failure probability of device performance degradation are derived, and the probability distribution of the device's remaining life is given, enabling device remaining life prediction.

[0108] The distribution of the remaining life of the equipment under discrete random shocks can help equipment managers better understand the operating conditions of the equipment and provide a reliable decision-making basis for scientific equipment management.

[0109] Throughout this specification, terms such as "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" refer to at least one embodiment or example described in conjunction with specific features, structures, materials, or characteristics. These specific features, structures, materials, or characteristics may be combined in appropriate manners in one or more embodiments or examples. Furthermore, a skilled artisan may combine and integrate different embodiments or examples described in this specification and their features, unless they are mutually incompatible.

[0110] The logic and / or steps shown in the flowcharts or otherwise described can be considered a sequence of executable instructions for implementing the logical functions. These instructions can be embodied in any computer-readable medium for use by an instruction execution system, device, or apparatus. These systems, devices, or apparatuses include processor systems or other systems capable of receiving and executing instructions.

[0111] The above embodiments have detailed the principles and implementation methods of the present invention, and illustrated its working principles using specific examples. These examples are intended only to facilitate understanding of the method and core concepts of the present invention. Furthermore, actual implementation methods and applications may vary depending on the principles of the present invention. Therefore, this specification should not be construed as limiting the present invention.

Claims

1. A method for estimating the remaining life of equipment by integrating grey prediction modeling and discrete random impact damage, characterized in that: The specific steps include: S1. Collect degradation data of equipment performance indicators, analyze the degradation data and extract the characteristics of the degradation data; Step S1 is specifically as follows: The degradation data of equipment performance indicators are collected and sorted, and discrete time series are used to represent the performance degradation of the equipment after being subjected to discrete random shocks. Each discrete random shock is recorded as a moment; from the first moment to the n The equipment performance index degradation sequence at time is recorded as: ; At the same time, the cumulative degradation of the equipment performance indicators after each impact is calculated, and the formula is expressed as: ； in, express i The degradation of equipment performance indicators at each moment; S2. Based on the degradation data characteristics, find the equipment performance degradation characteristics and build a gray prediction model that integrates random shocks to estimate the remaining life of the equipment. Step S2 is specifically as follows: The cumulative degradation of the equipment performance index obtained in step S1 , analyze the growth trend of performance index degradation after discrete random shock, and construct a discrete random grey prediction model. The formula is expressed as: ； in, For the device The number of shocks received during the time period, It obeys a homogeneous Poisson process with the impact intensity parameter ; 、 are the two parameters to be solved; S3. Solve the parameters of the grey prediction model constructed in step S2, and solve the impact characteristics based on the parameters of the grey prediction model, including: the number of first impacts and the average degradation trajectory; then calculate the equipment failure probability; step S3 specifically includes: S31: The cumulative degradation of the equipment performance index obtained in step S1 , solve the parameters of the random discrete grey prediction model, and its calculation formula is as follows: ； in, ; ; S32. Based on the discrete random grey prediction model and the failure threshold of the equipment performance index, the first impact number T is solved. The specific calculation formula is: ； in, is the critical value of equipment performance indicator failure, is the initial value of the equipment performance index degradation, is the value function; S33. Based on the number of first impacts, the discrete random grey prediction model and the properties of the Poisson process, the average degradation trajectory of the equipment performance degradation index is solved. It is expressed as the mathematical expectation of the degradation amount, and the formula is expressed as follows: ； S34. According to the number of first-arrival shocks and the properties of the Poisson process, we can get The probability of receiving T random shocks is: ； In the time period The probability of equipment failure is: ； in, is the impact strength parameter; S4. Based on the impact characteristics, solve the probability distribution of the remaining life of the equipment, calculate the equipment reliability, and estimate the average remaining life of the equipment.

2. The method for estimating the remaining life of equipment by integrating grey prediction modeling and discrete random impact damage according to claim 1 is characterized in that: Step S4 specifically includes: S41. Calculate the distribution of the remaining life of the equipment based on the random discrete grey prediction model that follows the equipment performance degradation law. Gamma distribution , its probability density function is: ； Its reliability function is: ； S42. Distribution based on the remaining life of the equipment , calculate the mathematical expectation of the distribution: ； S43, according to the impact arrival interval time , estimate the impact strength parameters: ； For the device to be The first shock and the The time interval between shocks, The total number of times the device is impacted; S44. Substitute the estimated value of the impact intensity parameter calculated in step S43 and the number of first impacts calculated in step S32 into the expected distribution of the remaining life of the equipment to obtain the estimated value of the average remaining life of the equipment.

3. A device, characterized in that: The device comprises a memory and a processor, wherein: a memory for storing computer programs capable of running on the processor; A processor is configured to execute, when running the computer program, a method for estimating the remaining life of equipment that integrates grey prediction modeling and discrete random impact damage as described in any one of claims 1-2.

4. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement a method for estimating the remaining life of equipment that integrates gray prediction modeling and discrete random impact damage according to any one of claims 1 to 2 when executed.

Images