A smart meter residual life prediction method and system

Abstract

This invention provides a method and system for predicting the remaining lifetime of smart meters, relating to the field of smart meter technology. The method includes: acquiring operational data of the smart meter; calculating the coupling strength between different fault types of the smart meter; establishing a fault propagation equation describing the various fault types in the operational data based on the coupling strength; calculating the maximum Lyapunov exponent and Jacobian matrix of the fault propagation equation; generating the remaining lifetime probability density of the smart meter using first-arrival time theory based on the maximum Lyapunov exponent and Jacobian matrix; and outputting the remaining lifetime prediction interval of the smart meter based on the remaining lifetime probability density. This method overcomes the limitations of traditional linear models and single fault assumptions, realizes dynamic evolution modeling of multi-source coupled faults, and improves the accuracy and reliability of remaining lifetime prediction.

Description

Technical Field

[0001] This invention relates to the field of smart meter technology, and in particular to a method and system for predicting the remaining lifespan of a smart meter. Background Technology

[0002] A smart meter is an electrical device that integrates communication, metering, and control technologies. It can collect electricity consumption data in real time and remotely, and transmit the data to the power company. Compared with traditional meters, it has higher accuracy, automation, and remote monitoring capabilities, helping users and power companies manage electricity use more efficiently.

[0003] As smart meters age, their hardware and software may gradually deteriorate or malfunction, affecting the accuracy of electricity metering and the stability of the system. Accurately predicting the remaining lifespan of the meters can help power companies plan equipment maintenance or replacement in advance, reducing the risk of downtime and sudden failures, while simultaneously improving the reliability of power management and the user's electricity experience.

[0004] However, existing technologies for predicting the remaining lifespan of smart meters often neglect the mutual influence and coupling between various faults within the smart meter. They often rely solely on prediction models for a single fault type or make subjective analyses based on linear assumptions, making it difficult to accurately predict the remaining lifespan of smart meters. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method for predicting the remaining life of a smart meter, which can solve the technical problem that the existing technology usually ignores the mutual influence and coupling between various faults inside the smart meter in predicting the remaining life of the smart meter, and often relies solely on the prediction model of a certain fault type or subjective analysis based on linear assumptions, making it difficult to accurately predict the remaining life of the smart meter.

[0006] A first aspect of this invention provides a method for predicting the remaining lifespan of a smart meter, comprising:

[0007] S1: Obtain the operating data of the smart meter;

[0008] S2: Calculate the coupling strength between different fault types of smart meters;

[0009] S3: Based on the coupling strength, establish the fault propagation equations describing the various fault types in the operating data;

[0010] S4: Calculate the maximum Lyapunov exponent and Jacobian matrix of the fault propagation equation;

[0011] S5: Based on the maximum Lyapunov exponent and Jacobian matrix, the remaining lifetime probability density of smart meters is generated through first-arrival time theory.

[0012] S6: Output the remaining life prediction range of the smart meter based on the remaining life probability density.

[0013] A second aspect of this invention provides a smart meter remaining life prediction system, comprising: a processor and a memory;

[0014] The memory stores a program or instructions that can run on a processor, which, when executed by the processor, implement the steps of the smart meter remaining life prediction method as described in the first aspect.

[0015] A third aspect of the present invention provides a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the smart meter remaining life prediction method of the first aspect.

[0016] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0017] In this embodiment of the invention, by quantifying the dynamic coupling strength among multiple fault types and constructing a nonlinear fault propagation equation, nonlinear dynamics theory is introduced into smart meter lifetime prediction. The maximum Lyapunov exponent is used to characterize the chaotic properties of the system, and the Jacobian matrix is ​​combined to characterize the stability of the fault propagation path. Finally, based on the first-arrival time theory, a remaining lifetime probability density function conforming to the actual fault evolution law is generated, and the remaining lifetime prediction interval is output. This method overcomes the limitations of traditional linear models and single fault assumptions, realizes dynamic evolution modeling of multi-source coupled faults, and improves the accuracy and reliability of remaining lifetime prediction. Attached Figure Description

[0018] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0019] Figure 1 This is a flowchart illustrating a method for predicting the remaining lifespan of a smart meter according to an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of a smart meter remaining life prediction system provided in an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions 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. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] The remaining life prediction method for smart meters provided by the present invention will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.

[0023] Reference manual attached Figure 1 The diagram shows a flowchart of a method for predicting the remaining lifespan of a smart meter according to an embodiment of the present invention.

[0024] This invention provides a method for predicting the remaining lifespan of a smart meter, which may include the following steps:

[0025] S1: Obtain the operating data of the smart meter.

[0026] Among them, operational data refers to various data collected by smart meters during actual use.

[0027] In one possible implementation, the operational data includes power metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate.

[0028] It should be noted that these data reflect the operating accuracy, stability, and environmental adaptability of smart meters. Electricity metering error indicates the meter's accuracy, voltage fluctuation variance reflects the stability of the power grid, ambient temperature gradient reveals the impact of temperature changes on meter performance, and communication error rate indicates the reliability of data transmission. Combining these data helps to comprehensively assess the meter's operating status and potential failure risks.

[0029] S2: Calculate the coupling strength between different fault types of smart meters.

[0030] Here, fault type refers to the different types of faults that may occur during the operation of a smart meter. Coupling strength is an indicator that measures the degree of mutual influence between different fault types. It represents the extent to which the occurrence of one fault type affects the occurrence of other fault types. High coupling strength means that one fault may cause other faults to occur, thereby accelerating the damage to the equipment.

[0031] By calculating the coupling strength between different fault types in smart meters, the interrelationships between faults can be revealed, thus enabling more accurate prediction of equipment failure trends. This process helps to comprehensively analyze potential fault risks in meters, take preventative measures in advance, avoid cascading reactions leading to more serious faults or outages, and improve the reliability of power management.

[0032] In one possible implementation, S2 specifically includes:

[0033] S201: Obtain historical operating data of smart meters under different fault types. The fault types include power metering module fault, power supply module fault, and communication module fault. The historical operating data includes multiple power metering errors, multiple voltage fluctuation variances, multiple ambient temperature gradients, and multiple communication error rates of the same type of smart meter.

[0034] S202: Extract fault characteristics of different fault types from historical operating data.

[0035] In one possible implementation, S202 specifically includes:

[0036] S2021: Calculate the mean of electricity metering error, the mean of voltage fluctuation variance, the mean of ambient temperature gradient, and the mean of communication bit error rate in historical operation data.

[0037] S2022: The mean values ​​of power metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate are normalized using a zero-one normalization method.

[0038] The zero-one normalization method specifically involves subtracting the minimum value of the same variable from the mean values ​​of the power metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate corresponding to the same fault type. Then, the result of the subtraction is divided by the difference between the maximum and minimum values ​​of the same variable, which yields the normalized mean values ​​of the power metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate.

[0039] S2023: The normalized mean values ​​of power metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate for the same fault type are combined to obtain a four-dimensional vector, i.e., the fault characteristics.

[0040] It is understandable that the fault characteristics are specifically a four-dimensional vector including the normalized mean elements of the power metering error, the mean elements of the voltage fluctuation variance, the mean elements of the ambient temperature gradient, and the mean elements of the communication bit error rate.

[0041] It should be noted that statistical calculations and normalization were performed on historical operational data regarding electricity metering errors, voltage fluctuation variance, ambient temperature gradients, and communication bit error rates to ensure comparability across different data dimensions. First, the mean of each data feature was calculated. Then, a zero-one normalization method was used, subtracting the minimum value from each variable and dividing by the difference between the maximum and minimum values ​​to obtain normalized data within a uniform range. Finally, these normalized data were combined into a four-dimensional vector, serving as the fault feature for each fault type. This method effectively standardized different data dimensions, improved the relative importance of each feature in subsequent analysis, and made the fusion of multi-dimensional data more accurate.

[0042] S203: Perform singular value decomposition on the fault features corresponding to each fault type to obtain the singular value matrix.

[0043] Specifically, the singular value decomposition (SVD) process involves first decomposing the fault feature matrix for each fault type. Specifically, the fault feature matrix is ​​decomposed into a product of three matrices: a left singular matrix, a diagonal matrix containing singular values, and a right singular matrix. The left singular matrix represents the correlation between fault features, the right singular matrix reveals the projection of features in different directions, and the singular values ​​in the diagonal matrix represent the importance and weight of each feature. This decomposition method extracts the most critical features from the original data, removes redundant information, simplifies the data structure, and improves the accuracy and efficiency of subsequent analysis.

[0044] S204: Combine the singular value matrix to calculate the outer product between each pair of fault features in order to capture the synergistic effect between any two fault types.

[0045] S205: Calculate the gradient of each fault feature to capture the changing trend of the corresponding fault feature.

[0046] The fault feature gradient calculation process involves point-by-point differentiation of each fault feature to capture its changing trend over time or other relevant variables. The gradient reflects the rate and direction of change of the feature value, i.e., the sensitivity of the fault feature to changes in time or state. By calculating the gradient, the trend of fault feature changes can be revealed, and the potential risk of fault development can be determined.

[0047] S206: Combine the outer product and gradient to calculate the coupling strength.

[0048] It should be noted that this process comprehensively quantifies the coupling strength of different fault types through multi-dimensional data fusion and dynamic feature extraction. Specifically, firstly, fault features of power metering, power supply, and communication modules are extracted based on historical operating data, and singular value decomposition is used to eliminate redundancy while retaining key dynamic information. Secondly, the nonlinear synergistic effect between fault features is captured through outer product operations, and gradient analysis is combined to reveal the fault evolution trend. Finally, the static synergistic relationship and dynamic change characteristics are integrated to construct a more comprehensive coupling strength calculation model, effectively solving the problem of insufficient representation of coupling relationships caused by single feature analysis in traditional methods, and providing more accurate input parameters for subsequent fault propagation modeling.

[0049] The specific formula for calculating coupling strength is as follows:

[0050] .

[0051] in, This represents the coupling strength between the k-th fault type and the l-th fault type. This represents the natural exponential function. These represent the maximum singular values ​​in the singular value matrix corresponding to the k-th fault type. This represents the maximum singular value in the singular value matrix corresponding to the l-th fault type. and Let L1 norm and L2 norm be represented respectively. and These represent the fault characteristics of the k-th fault type and the l-th fault type, respectively. This represents the outer product operation. Indicates calculation gradient and gradient The inner product between them.

[0052] It should be noted that this coupling strength formula quantifies the mutual influence between different fault types through multiple factors. First, the outer product operation is used to capture the synergistic effect between fault features, and the L1 norm and L2 norm are combined to measure the differences and gradient changes between features. By calculating the inner product of the fault feature gradients, the fault evolution trend is further revealed, and it is modulated by the natural exponential function. The maximum singular value is used to weigh the degree of influence of different fault types, and finally, the accurate coupling strength is obtained, providing a reliable foundation for subsequent fault propagation analysis.

[0053] S3: Based on the coupling strength, establish the fault propagation equations describing the various fault types in the operating data.

[0054] The fault propagation equation is a mathematical model describing the mutual influence and propagation mechanisms between different fault types. By combining coupling strength to establish the fault propagation equation, the propagation process of faults in smart meters can be comprehensively simulated and analyzed. This allows the prediction of fault occurrence to go beyond a single fault type, taking into account the complex interrelationships between multiple faults. Through this equation, the dynamic process of fault development can be grasped more accurately, improving the accuracy and reliability of remaining lifetime prediction.

[0055] In one possible implementation, S3 specifically includes:

[0056] S301: Based on the fault threshold of the smart meter, the operating data is converted into a degradation degree, which includes the degradation degree of the power metering module, the degradation degree of the power supply module, and the degradation degree of the communication module.

[0057] Alternatively, the specific fault thresholds can be found in the industry standards or component manuals for each module of the smart meter.

[0058] It is understandable that the degradation degree includes the degradation degree of the power metering module, the power supply module, and the communication module. Specifically, the degradation degree of the power metering module is the ratio of the power metering error in the operating data to the power metering error fault threshold; the degradation degree of the power supply module is specifically the ratio of the voltage fluctuation variance in the operating data to the voltage fluctuation variance fault threshold; and the degradation degree of the communication module is specifically the ratio of the communication bit error rate in the operating data to the communication bit error rate fault threshold.

[0059] S302: Combine the degradation degree of the power metering module, the degradation degree of the power supply module, and the degradation degree of the communication module to obtain the degradation degree state vector of the smart meter.

[0060] S303: Based on the state vector of smart meter degradation and coupling strength, establish the fault propagation equation.

[0061] The specific formula for the fault propagation equation is as follows:

[0062] .

[0063] in, This represents the state vector of the smart meter's degradation level at time t. This represents the independent failure rate of each module in the experimentally calibrated smart meter. This represents a coupling matrix where the elements are the coupling strength between different fault types. This represents the Hadamard product operation.

[0064] Specifically, this process constructs a nonlinear fault propagation model by quantifying the degradation degree and coupling effect of the modules. Specifically, the ratios of power metering error, voltage fluctuation variance, and communication bit error rate to the fault threshold are converted into normalized degradation degree indices, forming a three-dimensional degradation degree state vector. Then, based on experimentally calibrated independent failure rate matrix and fault coupling matrix, a model containing linear degradation terms is established. Coupling terms with nonlinearity The propagation equation is derived, where the Hadamard product represents the nonlinear interaction of module degradation. This model reflects both the exponential decay characteristics of independent module failures and the dynamic capture of the collaborative evolution mechanism among multiple fault types through the coupling matrix, significantly improving the modeling accuracy of complex fault propagation paths in smart meters.

[0065] S4: Calculate the maximum Lyapunov exponent and Jacobian matrix of the fault propagation equation.

[0066] The maximum Lyapunov exponent describes the sensitivity of a system to fault propagation, i.e., the system's dependence on initial conditions. A larger Lyapunov exponent indicates that the system diverges rapidly in the face of small disturbances, potentially leading to fault propagation and unpredictable behavior. The Jacobian matrix is ​​a matrix describing the relationships between the system's variables. In the fault propagation equation, the Jacobian matrix represents the influence of each fault type on other fault types, helping to analyze the system's stability and how faults propagate within the system.

[0067] By calculating the maximum Lyapunov exponent and the Jacobian matrix, the stability and sensitivity of the fault propagation system can be evaluated. This provides a quantitative mathematical basis for predicting fault development trends and assessing the health status of smart meters, accurately grasping the potential risks of fault propagation, helping to provide early warnings and plan equipment maintenance, and reducing the occurrence of sudden failures.

[0068] In one possible implementation, the Jacobian matrix is ​​specifically the result of taking the partial derivatives of each term in the fault propagation equation with respect to the state vector of the smart meter's degradation degree. The maximum Lyapunov exponent is specifically the ratio of the natural logarithm of the quotient between the magnitude of the disturbance vector at time t and the magnitude of the disturbance vector at the initial time, under the constraint of an infinite time region, to the value at time t.

[0069] The formula for calculating the Jacobian matrix is ​​as follows:

[0070] .

[0071] in, Let represent the Jacobian matrix at time t. This indicates the partial derivative.

[0072] The formula for calculating the maximum Lyapunov exponent is as follows:

[0073] .

[0074] in, This represents the maximum Lyapunov index. This indicates that time t approaches infinity, and ln represents the natural logarithm. It represents a small change. This represents the magnitude of the perturbation vector at the initial moment. Let t represent the magnitude of the perturbation vector at time t.

[0075] Specifically, the Jacobian matrix describes the sensitivity of the smart meter's degradation state vector to changes in terms of the fault propagation equation. More specifically, it is the result of taking the partial derivative of the degradation state vector, representing the local changes in the system at a specific moment. This process captures the nonlinear evolution of the system state through the effects of changes in degradation degree and coupling strength. The maximum Lyapunov exponent measures the system's stability; by calculating the growth rate of the disturbance vector over time, it determines the system's sensitivity to initial disturbances. A large exponent indicates that the system is highly sensitive to fault changes and prone to instability, thus affecting the meter's predicted lifetime. This provides key parameters for subsequent lifetime probability density modeling based on first-arrival time theory, significantly improving the accuracy and reliability of remaining lifetime prediction.

[0076] S5: Based on the maximum Lyapunov exponent and Jacobian matrix, the remaining lifetime probability density of smart meters is generated through first-arrival time theory.

[0077] First-arrival time theory (FATH) is a method for analyzing stochastic processes, typically used to study the time required for a system to reach a specific state from an initial state. In predicting the remaining lifetime of smart meters, FTH is used to estimate the probability of a fault occurring for the first time at a future point in time, i.e., the time it takes for the system to transition from normal operation to a fault state. The remaining lifetime probability density function is a probability distribution function describing the probability of a smart meter failing within a certain future timeframe. It reflects the probability distribution of the remaining time from the current moment until a fault occurs, helping to predict its remaining lifetime.

[0078] By combining the maximum Lyapunov exponent, the Jacobian matrix, and first-arrival time theory, a probability density distribution of the remaining lifetime of smart meters can be generated. This process not only considers the fault propagation characteristics of the meters but also fully reflects the different fault risks that the equipment may face in the future, providing a more accurate and detailed prediction of the remaining lifetime. This helps power companies develop more reasonable maintenance plans and risk management strategies, extending the service life of equipment.

[0079] In one possible implementation, S5 specifically includes:

[0080] S501: Establish the Lyapunov equation based on the Jacobian matrix, and solve the Lyapunov equation to obtain the positive definite matrix.

[0081] The Lyapunov equations are as follows:

[0082] .

[0083] in, Let T denote the Jacobian matrix at time t, and let the subscript T denote the transpose. Describes a positive definite matrix. Represents the identity matrix.

[0084] It should be noted that by establishing the Lyapunov equations and solving for the positive definite matrix, the stability of the system can be effectively evaluated. This method, by analyzing the properties of the Jacobian matrix, reveals the sensitivity of the system to state changes, helping to accurately capture the dynamic behavior of fault propagation, thereby improving the accuracy and reliability of remaining lifetime prediction.

[0085] S502: Combine the positive definite matrix to establish an energy function for the state vector of the smart meter's degradation degree. Specifically, the energy function is the product of the state vector of the smart meter's degradation degree, the positive definite matrix, and the transpose of the state vector of the smart meter's degradation degree.

[0086] The energy function is specifically expressed in the following formula:

[0087] .

[0088] in, Represents the state vector of the smart meter's degradation degree at time t. The energy function value.

[0089] It should be noted that by establishing an energy function, the energy changes of a smart meter's degradation state over time can be quantified, thereby revealing its degradation process. Using a positive definite matrix to calculate the energy function helps to accurately describe the system's stability and trends, thus more accurately predicting the meter's remaining lifespan and avoiding the risks associated with potential failures.

[0090] S503: Calculate the critical smart meter degradation state vector when the maximum Lyapunov exponent is zero, and calculate the critical energy function value based on the critical smart meter degradation state vector.

[0091] The calculation method for the critical degradation state vector of a smart meter is as follows: First, establish the functional relationship between the maximum Lyapunov exponent and the smart meter degradation state vector: ,in, This represents taking the real part of the eigenvalue. Then, a numerical iterative method, such as Newton's method, is used to find the eigenvalue that satisfies... The state vector that is zero is the critical smart meter degradation state vector. By substituting the critical smart meter degradation state vector into the energy function, the critical energy function value can be obtained. .

[0092] It should be noted that by calculating the critical degradation state vector when the maximum Lyapunov exponent is zero, the critical point of the smart meter system can be accurately captured. This helps identify the critical state of system stability, provides an important benchmark for remaining lifetime prediction, ensures the advance and accuracy of fault prediction, and effectively avoids premature equipment failure.

[0093] S504: Calculate the remaining lifetime probability density based on the first arrival time theory by combining the energy function and the critical energy function value.

[0094] The specific formula for calculating the remaining lifetime probability density is as follows:

[0095] .

[0096] in, Let t represent the probability density of smart meter failure at time t, i.e., the probability density of remaining lifetime. express gradient, Represents pi (π). This represents the critical energy function value. This represents the cumulative variance of environmental disturbances obtained from statistics.

[0097] The cumulative variance of environmental disturbances refers to the cumulative variance of random fluctuations caused by external environmental factors during the operation of the smart meter, which can be specifically calculated by temperature. By combining the energy function and critical energy function values, and using the first-arrival time theory to calculate the remaining lifetime probability density, the failure time of the smart meter can be predicted more accurately. This method considers system state changes, the impact of environmental disturbances, and degradation trends, providing a reliable estimate of remaining lifetime.

[0098] Specifically, this process constructs an energy function using Lyapunov stability theory and combines it with first-arrival time theory to achieve probabilistic modeling of the remaining lifetime. Specifically, it obtains a positive definite matrix by solving the Lyapunov equations and constructs a quadratic form function characterizing the system's energy state. The system instability threshold and corresponding critical energy are determined using the critical condition that the maximum Lyapunov exponent is zero. Finally, based on the energy gradient and the statistical characteristics of environmental disturbances, the first-arrival time probability density of the system from the current state to the critical state is quantified using a Gaussian distribution. The exponential term characterizes the statistical distribution of energy deviation from the critical value, and the gradient term reflects the rate of evolution of the system state towards the critical point. This enables nonlinear prediction of the remaining lifespan of smart meters, significantly improving the reliability of lifespan prediction in multi-fault coupled scenarios.

[0099] S6: Output the remaining life prediction range of the smart meter based on the remaining life probability density.

[0100] The remaining lifetime prediction interval refers to the range of possible failure times calculated based on the current state of the smart meter and its remaining lifetime probability density.

[0101] In one possible implementation, S6 specifically includes:

[0102] S601: Obtain the preset significance level for the remaining lifetime prediction interval as false.

[0103] It should be noted that those skilled in the art can set the preset salience level according to actual needs, and this invention does not limit this.

[0104] S602: Calculate the cumulative remaining lifetime probability density.

[0105] The cumulative remaining lifetime probability density is calculated using the cumulative remaining lifetime probability density function, which is as follows:

[0106] .

[0107] in, This represents the cumulative remaining lifetime probability density of the smart meter at time t. This represents the time integral variable.

[0108] S603: Calculate the remaining lifetime prediction interval at a preset significance level using the inverse function of the cumulative remaining lifetime probability density function.

[0109] The specific formula for calculating the remaining lifetime prediction range is as follows:

[0110] .

[0111] in, Indicates the pre-set significance level The remaining life prediction range at that time This represents the inverse function of the remaining lifetime probability density function. and Representing the significance levels respectively. The relevant lower limit and upper limit of remaining lifetime prediction.

[0112] Specifically, the cumulative remaining lifetime probability density function is obtained by integrating the remaining lifetime probability density function, representing the probability that the smart meter will malfunction before time t. The inverse function represents finding the corresponding time point, i.e., the time point where the cumulative probability density equals that probability, given a probability. Therefore, the inverse function can convert a given probability value into a specific time value of the remaining lifetime. At a significance level of... At that time, the lower limit of the remaining lifetime is obtained through the inverse function. This is calculated. It represents the lower bound of the remaining lifetime under worst-case conditions, meaning the time probability of failure is... The upper limit of remaining lifespan is determined by the inverse function. This is calculated to represent the optimal upper bound of the remaining lifetime, implying that the time probability of failure is... The remaining life prediction interval is represented at a confidence level of 1- Below is the possible range of smart meter failure times. For example, if... Taking 0.05, the lower and upper limits of the inverse function are 500 days and 1000 days respectively. Therefore, the remaining lifetime prediction interval means that there is a 2.5% probability of failure within 500 days and a 97.5% probability of failure within 1000 days.

[0113] Specifically, this process determines the remaining lifetime prediction interval by calculating the cumulative probability density function of the remaining lifetime. First, the cumulative probability density function is calculated based on the remaining lifetime probability density. Then, using its inverse function, the upper and lower limits of the remaining lifetime are determined at a set significance level. This method provides a confidence interval, representing the possible time range for smart meter failure at a specific significance level, thus improving the accuracy of the prediction interval.

[0114] In practical applications, this process integrates multi-dimensional operational data with nonlinear dynamics theory to construct a dynamic fault propagation model and generate a high-precision lifetime probability density prediction interval. The specific process includes: first, collecting operational data such as electricity metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate. Fault features are extracted from historical data through normalization and singular value decomposition, and the coupling strength between different fault types is quantified by combining outer product operations and gradient analysis. Subsequently, a fault propagation equation containing linear degradation terms and nonlinear coupling terms is established, and the Jacobian matrix and maximum Lyapunov exponent are calculated to characterize system stability. Further, the positive definite matrix is ​​solved using the Lyapunov equation to construct the energy function and determine the critical degradation state. Finally, the remaining lifetime probability density function is generated based on the first-arrival time theory. Finally, the confidence interval is calculated using the cumulative probability density inverse function, outputting a statistically reliable remaining lifetime prediction interval. The introduction of the Lyapunov exponent and the energy function for critical state determination enables transitional modeling from deterministic degradation to probabilistic failure, effectively solving the problems of single fault assumptions, static analysis, and large dispersion of prediction results in existing technologies.

[0115] In this embodiment of the invention, by quantifying the dynamic coupling strength among multiple fault types and constructing a nonlinear fault propagation equation, nonlinear dynamics theory is introduced into smart meter lifetime prediction. The maximum Lyapunov exponent is used to characterize the chaotic properties of the system, and the Jacobian matrix is ​​combined to characterize the stability of the fault propagation path. Finally, based on the first-arrival time theory, a remaining lifetime probability density function conforming to the actual fault evolution law is generated, and the remaining lifetime prediction interval is output. This method overcomes the limitations of traditional linear models and single fault assumptions, realizes dynamic evolution modeling of multi-source coupled faults, and improves the accuracy and reliability of remaining lifetime prediction.

[0116] Reference manual attached Figure 2 The diagram shows a structural schematic of a smart meter remaining life prediction system provided by an embodiment of the present invention.

[0117] This invention provides a smart meter remaining life prediction system 20, including: a processor 201 and a memory 202;

[0118] The memory 202 stores programs or instructions that can run on the processor 201. When the program or instructions are executed by the processor 201, they implement the steps of the above-described method for predicting the remaining lifespan of a smart meter and achieve the same technical effect. To avoid repetition, the present invention will not elaborate further.

[0119] It should be understood that the processor 201 in this embodiment of the invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0120] It should also be understood that the memory 202 in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DR RAM).

[0121] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0122] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0123] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0124] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0125] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0126] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0127] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0128] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] This invention provides a readable storage medium comprising: storing a program or instructions on the readable storage medium, wherein when the program or instructions are executed by a processor, the program or instructions implement the steps of the above-described method for predicting the remaining lifespan of a smart meter, and can achieve the same technical effect. To avoid repetition, this invention will not elaborate further.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting the remaining lifespan of a smart meter, characterized in that, include: S1: Obtain the operating data of the smart meter; S2: Calculate the coupling strength between different fault types of the smart meter; S3: Based on the coupling strength, establish a fault propagation equation describing the various fault types of the operating data; S4: Calculate the maximum Lyapunov exponent and Jacobian matrix of the fault propagation equation; S5: Based on the maximum Lyapunov exponent and the Jacobian matrix, the remaining lifetime probability density of the smart meter is generated using the first-arrival time theory; S6: Output the remaining lifetime prediction range of the smart meter based on the remaining lifetime probability density; Specifically, S2 includes: S201: Obtain historical operating data of the smart meter under different fault types, wherein the fault types include power metering module fault, power supply module fault and communication module fault, and the historical operating data includes multiple power metering errors, multiple voltage fluctuation variances, multiple ambient temperature gradients and multiple communication bit error rates of the same type of smart meter; S202: Extract fault characteristics of different fault types from the historical operation data; S203: Perform singular value decomposition on the fault features corresponding to each fault type to obtain a singular value matrix; S204: Combine the singular value matrix to calculate the outer product between any two of the fault features to capture the synergistic effect between any two of the fault types; S205: Calculate the gradient of each of the fault features to capture the changing trend of the corresponding fault features; S206: Calculate the coupling strength by combining the outer product and the gradient; The specific formula for calculating the coupling strength is as follows: in, This represents the coupling strength between the k-th fault type and the l-th fault type. This represents the natural exponential function. These represent the maximum singular values ​​in the singular value matrix corresponding to the k-th fault type. This represents the maximum singular value in the singular value matrix corresponding to the l-th fault type. and Let L1 norm and L2 norm be represented respectively. and These represent the fault characteristics of the k-th fault type and the l-th fault type, respectively. This represents the outer product operation. Indicates calculation gradient and gradient The inner product between; S5 specifically includes: S501: Establish the Lyapunov equation based on the Jacobian matrix, and solve the Lyapunov equation to obtain a positive definite matrix; The Lyapunov equations are as follows: in, Let T represent the Jacobian matrix at time t, where the subscript T indicates the transpose. Describes a positive definite matrix. Represents the identity matrix; S502: Combining the positive definite matrix, establish an energy function for the degradation state vector of the smart meter, wherein the energy function is specifically the product of the degradation state vector of the smart meter, the positive definite matrix, and the transpose of the degradation state vector of the smart meter; S503: Calculate the critical smart meter degradation state vector when the maximum Lyapunov exponent is zero, and calculate the critical energy function value based on the critical smart meter degradation state vector; S504: Combining the energy function and the critical energy function value, calculate the remaining lifetime probability density based on the first arrival time theory; The specific formula for calculating the remaining lifetime probability density is as follows: in, Let t represent the probability density of smart meter failure at time t, i.e., the probability density of remaining lifetime. express gradient, Represents pi (π). This represents the critical energy function value. This represents the cumulative variance of environmental disturbances obtained from statistics.

2. The method for predicting the remaining lifespan of a smart meter according to claim 1, characterized in that, The operational data includes power metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate.

3. The method for predicting the remaining lifespan of a smart meter according to claim 1, characterized in that, S202 specifically includes: S2021: Calculate the mean of power metering error, the mean of voltage fluctuation variance, the mean of ambient temperature gradient, and the mean of communication bit error rate in the historical operating data; S2022: Normalize the mean value of the power metering error, the mean value of the voltage fluctuation variance, the mean value of the ambient temperature gradient, and the mean value of the communication bit error rate using a zero-one normalization method; S2023: Combine the normalized mean values ​​of power metering error, voltage fluctuation variance, ambient temperature gradient, and communication bit error rate for the same fault type to obtain a four-dimensional vector, which is the fault feature.

4. The method for predicting the remaining lifespan of a smart meter according to claim 1, characterized in that, S3 specifically includes: S301: Based on the fault threshold of the smart meter, the operating data is converted into a degradation degree, wherein the degradation degree includes the degradation degree of the power metering module, the degradation degree of the power supply module, and the degradation degree of the communication module; S302: Combine the degradation degree of the power metering module, the degradation degree of the power supply module, and the degradation degree of the communication module to obtain the smart meter degradation degree state vector; S303: Based on the degradation state vector of the smart meter and the coupling strength, establish the fault propagation equation.

5. The method for predicting the remaining lifespan of a smart meter according to claim 4, characterized in that, The Jacobian matrix is ​​specifically the result of taking the partial derivative of each term in the fault propagation equation with respect to the state vector of the smart meter's degradation degree; the maximum Lyapunov exponent is specifically the ratio of the natural logarithm of the quotient between the magnitude of the disturbance vector at time t and the magnitude of the disturbance vector at the initial time to the value at time t, under the constraint of an infinite time region.

6. The method for predicting the remaining lifespan of a smart meter according to claim 1, characterized in that, S6 specifically includes: S601: Obtain the preset significance level for the remaining lifetime prediction interval as false; S602: Calculate the cumulative remaining lifetime probability density of the remaining lifetime probability density; S603: Calculate the remaining lifetime prediction interval at the preset significance level using the inverse function of the cumulative remaining lifetime probability density function.

7. A system for predicting the remaining lifespan of a smart meter, characterized in that, include: Processor and memory; The memory stores programs or instructions that can run on the processor, which, when executed by the processor, implement the steps of the smart meter remaining life prediction method as described in any one of claims 1 to 6.

8. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the smart meter remaining life prediction method as described in any one of claims 1 to 6.

Images