A method and device for predicting damage to electronic equipment in a complex post-disaster environment

Abstract

The present application provides a method and device for predicting damage of electronic equipment in a complex post-disaster environment. An accelerated degradation model is constructed using a power-law environmental comprehensive stress model and a Wiener degradation model of electronic equipment. The accelerated degradation model characterizes the accelerated degradation process of electronic equipment under environmental comprehensive stress. Based on the accelerated degradation process, an average life prediction model integrating the power-law environmental comprehensive stress model and the data-driven model is constructed. Based on the historical accelerated degradation data set of the electronic equipment, the target estimated values ​​of the undetermined parameters of the accelerated degradation model are solved based on the particle swarm algorithm (PSO). Based on the target estimated values, the damage prediction value of the electronic equipment under environmental comprehensive stress is determined based on the average life prediction model. The damage prediction value includes the average life prediction value. Taking into account the influence of various environmental factors on life prediction in a damaging environment, the accelerated degradation model optimized by the particle swarm algorithm is used to improve the accuracy and reliability of the average life prediction model.

Description

Technical Field

[0001] The present application relates to the field of electronic equipment damage prediction, and in particular to a method and device for predicting electronic equipment damage in a complex post-disaster environment. Background Art

[0002] Smoke environments are complex. Factors such as ambient temperature, humidity, smoke composition, smoke concentration, and operating current can all affect the reliability of electronic equipment, accelerating failure and shortening its service life. In fire incidents, timely rescue and maintenance of electronic equipment are often impossible. Without backup equipment, electronic equipment must continue to operate in a damaging environment to maintain critical functions. Therefore, a method is needed to accurately predict damage to electronic equipment after complex fires.

[0003] Adaptive prediction methods for electronic equipment damage under the influence of fire smoke have been proposed. However, the influence of multiple factors is often not simply additive but rather synergistically coupled. Current prediction methods ignore the complex influence and synergistic coupling of multiple factors in a damaging environment, making it difficult to accurately reflect the actual damage to electronic equipment caused by these coupled factors. Consequently, current prediction methods have low accuracy.

[0004] Therefore, there is an urgent need to develop a prediction method for electronic equipment damage under the coupling of multiple factors after a fire, which can accurately predict the damage of electronic equipment in a complex environment after a fire and provide guidance for post-disaster assessment. Summary of the Invention

[0005] In view of this, the present application provides a method and device for predicting damage to electronic equipment in a complex post-disaster environment, which is used to accurately predict the damage of electronic equipment under the influence of multiple factors in a damaging environment. The accelerated degradation model optimized by the particle swarm algorithm can more accurately consider the comprehensive impact of multiple environmental factors (such as temperature, humidity, fire smoke components, fire smoke concentration, working current, etc.) on the damage, thereby improving the accuracy and reliability of the average life prediction model.

[0006] To solve the above problems, the technical solutions provided by this application are as follows:

[0007] In a first aspect, an embodiment of the present application provides a method for predicting damage to electronic equipment in a complex post-disaster environment, the method comprising:

[0008] Acquire a damage data set of an electronic device, wherein the damage data set includes accelerated degradation parameters of the electronic device under a preset damaging environment;

[0009] Constructing an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and the Wiener degradation model of the electronic device, wherein the accelerated degradation model characterizes the accelerated degradation process of the electronic device under the environmental comprehensive stress;

[0010] Based on the accelerated degradation process of the electronic device, constructing an average life prediction model for the electronic device that integrates the power-law environmental comprehensive stress model and the data-driven model;

[0011] According to the historical accelerated degradation data set, optimizing the undetermined parameters of the accelerated degradation model based on a particle swarm algorithm (PSO), wherein the particle swarm algorithm is used to determine target estimated values ​​of the undetermined parameters in the accelerated degradation model by performing a global search in a preset parameter space;

[0012] According to the target estimated values ​​of the undetermined parameters in the accelerated degradation model, the damage prediction value of the electronic device under the environmental comprehensive stress is determined based on the average life prediction model, and the damage prediction value includes the average life prediction value.

[0013] Optionally, the accelerated degradation parameters of the electronic device under a preset fire damage environment may include at least one of the following: on-resistance change rate, leakage current, voltage, junction temperature, etc.

[0014] Alternatively, a data-driven model is one that builds a model by learning from data rather than relying on traditional physical principles or assumptions. The Wiener degradation process primarily relies on a damage dataset of the electronic device, from which the data-driven model is constructed.

[0015] Optionally, constructing the accelerated degradation model of the electronic device by using the power-law environmental comprehensive stress model and the Wiener degradation model of the electronic device includes:

[0016] The degradation path of the electronic device is simulated using the Wiener degradation model of the electronic device, and the drift parameter μ and diffusion parameter σ of the Wiener degradation model are associated with the environmental comprehensive stress using the power-law environmental comprehensive stress model to construct an accelerated degradation model of the electronic device.

[0017] Optionally, the drift parameter μ(T,Q i ) is expressed as Diffusion parameter σ2(T,Q i ) is expressed as The accelerated degradation model of the electronic device is: The environmental comprehensive stress includes temperature stress T and non-thermal stress Q, wherein the non-thermal stress can be one or more of humidity stress, current stress, mechanical stress and electromagnetic stress, and n represents the number of non-thermal stresses; Y(t, S) is the accelerated degradation state of the electronic device under the action of environmental comprehensive stress at time t, N represents the normal distribution, a1, a2, b i , c1 and β are unknown parameters.

[0018] Optionally, the optimizing the undetermined parameters of the accelerated degradation model based on the historical accelerated degradation dataset and a particle swarm algorithm (PSO) is performed, wherein the particle swarm algorithm is configured to perform a global search in a preset parameter space and determine target estimated values ​​of the undetermined parameters in the accelerated degradation model, including:

[0019] Initializing a particle swarm according to the particle swarm algorithm (PSO), wherein the initialized particle swarm includes: the number of individuals in the particle swarm, the maximum number of iterations, and the search range of each particle, wherein each particle in the particle swarm represents an undetermined parameter in the accelerated degradation model, and the position and velocity of the particle respectively represent the value and change trend of the undetermined parameter. During initialization, the initial position and velocity of the particle are randomly distributed within a preset parameter space;

[0020] Obtaining a fitness function, wherein the fitness function is used to evaluate the degree of proximity between the undetermined parameter represented by each particle and the actual accelerated degradation data;

[0021] The speed and position of each particle in the particle swarm are searched and updated to iteratively approach the global optimal solution. In each iteration, the speed and position of the particle are updated according to the particle update rule to guide the particle swarm to move towards the global optimal position. In each update, the speed and position of the particle are affected by the current particle's own historical best position, the global best position, and random factors, ensuring that the particle swarm performs an effective global search in the search space.

[0022] After each iteration, the fitness of each particle is evaluated using the fitness function by using the historical accelerated degradation data set and the prediction error of the accelerated degradation model to determine whether the undetermined parameters of the accelerated degradation model corresponding to the current position of each particle can effectively fit the historical accelerated degradation data;

[0023] After reaching a predetermined maximum number of iterations or satisfying a convergence condition, target estimated values ​​of undetermined parameters of the optimized accelerated degradation model are output.

[0024] Optionally, based on the accelerated degradation process of the electronic device, a prediction model for the average life of the electronic device is constructed by integrating the power-law environmental comprehensive stress model and a data-driven model, wherein the data-driven model is generated based on a cumulative distribution function, including:

[0025] The power-law environmental comprehensive stress model and the data-driven model are integrated to construct an average life prediction model for the electronic device. The average life prediction model is: Among them, γ is the average predicted life expectancy, F γ (t; μ, σ, D) is the cumulative distribution function at time t, μ is the target drift parameter of the accelerated degradation model, σ 2 is the target diffusion parameter of the accelerated degradation model, and D is the preset failure threshold; the target drift parameter μ and the target diffusion parameter σ of the accelerated degradation model are both related to the environmental stress, and the relationship between the target drift parameter μ and the target diffusion parameter σ and the environmental stress is characterized by the power-law environmental comprehensive stress model.

[0026] In a second aspect, an embodiment of the present application provides a device for predicting damage to electronic equipment in a complex post-disaster environment, the device comprising:

[0027] an acquisition module, configured to acquire a damage dataset of an electronic device, wherein the damage dataset includes accelerated degradation parameters of the electronic device under a preset damaging environment;

[0028] A construction module is used to construct an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and a Wiener degradation model of the electronic device, wherein the accelerated degradation model characterizes the accelerated degradation process of the electronic device under the environmental comprehensive stress;

[0029] The construction module is further configured to construct, based on the accelerated degradation process of the electronic device, an average life prediction model for the electronic device that integrates the power-law environmental comprehensive stress model and a data-driven model, wherein the data-driven model is generated based on a cumulative distribution function;

[0030] a solution module, configured to optimize the undetermined parameters of the accelerated degradation model based on the historical accelerated degradation data set and a particle swarm algorithm (PSO), wherein the particle swarm algorithm is configured to determine target estimated values ​​of the undetermined parameters in the accelerated degradation model by performing a global search within a preset parameter space;

[0031] A prediction module is used to determine the damage prediction value of the electronic device under environmental comprehensive stress based on the average life prediction model according to the target estimated value of the undetermined parameter in the accelerated degradation model, wherein the damage prediction value includes the average life prediction value.

[0032] Optionally, the building block is specifically used to:

[0033] The degradation path of the electronic device is simulated using the Wiener degradation model of the electronic device, and the drift parameter μ and diffusion parameter σ of the Wiener degradation model are associated with the environmental comprehensive stress using the power-law environmental comprehensive stress model to construct an accelerated degradation model of the electronic device.

[0034] Optionally, the drift parameter μ(T,Q i ) is expressed as Diffusion parameter σ 2 (T,Q i ) is expressed as The accelerated degradation model of the electronic device is: The environmental comprehensive stress includes temperature stress T and non-thermal stress Q, wherein the non-thermal stress can be one or more of humidity stress, current stress, mechanical stress and electromagnetic stress, and n represents the number of non-thermal stresses; Y(t, S) is the accelerated degradation state of the electronic device under the action of environmental comprehensive stress at time t, N represents the normal distribution, a1, a2, b i , c1 and β are unknown parameters.

[0035] Optionally, the solution module includes: an initialization submodule, an acquisition submodule, a search and update submodule, an evaluation submodule and an output submodule;

[0036] An initialization submodule is used to initialize a particle swarm according to the particle swarm algorithm (PSO). The initialized particle swarm includes: the number of individuals in the particle swarm, the maximum number of iterations, and the search range of each particle. Each particle in the particle swarm represents an undetermined parameter in the accelerated degradation model. The position and velocity of the particle represent the value and change trend of the undetermined parameter, respectively. During initialization, the initial position and velocity of the particle are randomly distributed within a preset parameter space.

[0037] An acquisition submodule, used for acquiring a fitness function, wherein the fitness function is used to evaluate the degree of proximity between the undetermined parameter represented by each particle and the actual accelerated degradation data;

[0038] A search and update submodule is used to search and update the speed and position of each particle in the particle swarm to iteratively approach the global optimal solution. In each iteration, the speed and position of the particle are updated according to the particle update rule to guide the particle swarm to move towards the global optimal position. In each update, the speed and position of the particle are affected by the current particle's own historical best position, the global best position, and random factors, ensuring that the particle swarm performs an effective global search in the search space.

[0039] an evaluation submodule, configured to evaluate the fitness of each particle after each iteration by using the historical accelerated degradation data set and the prediction error of the accelerated degradation model using the fitness function to determine whether the undetermined parameters of the accelerated degradation model corresponding to the current position of each particle can effectively fit the historical accelerated degradation data;

[0040] The output submodule is used to output the target estimated values ​​of the undetermined parameters of the optimized accelerated degradation model after reaching a predetermined maximum number of iterations or satisfying a convergence condition.

[0041] Optionally, the construction module is further configured to construct, based on the accelerated degradation process of the electronic device, an average life prediction model for the electronic device that integrates the power-law environmental comprehensive stress model and a data-driven model, wherein the data-driven model is generated based on a cumulative distribution function and includes:

[0042] A fusion submodule is used to fuse the power-law environmental comprehensive stress model and the data-driven model to construct an average life prediction model for the electronic device. The average life prediction model is: Among them, γ is the average predicted life expectancy, F γ (t; μ, σ, D) is the cumulative distribution function at time t, μ is the target drift parameter of the accelerated degradation model, σ 2 is the target diffusion parameter of the accelerated degradation model, and D is the preset failure threshold; the target drift parameter μ and the target diffusion parameter σ of the accelerated degradation model are both related to the environmental stress, and the relationship between the target drift parameter μ and the target diffusion parameter σ and the environmental stress is characterized by the power-law environmental comprehensive stress model.

[0043] In a third aspect, an embodiment of the present application provides a device for predicting damage to electronic equipment in a complex post-disaster environment, the device comprising: a processor, a memory, and a system bus;

[0044] The processor and the memory are connected via the system bus;

[0045] The memory is used to store one or more programs, and the one or more programs include instructions. When the instructions are executed by the processor, the processor executes the method for predicting damage of electronic equipment in a complex post-disaster environment as described in the first aspect above.

[0046] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores instructions, and when the instructions are executed on a device, the device executes the method for predicting damage of electronic equipment in a complex post-disaster environment as described in the first aspect above.

[0047] It can be seen that this application has the following beneficial effects:

[0048] The present application provides a method for predicting damage to electronic devices in a complex post-disaster environment. When executing the method, a damage dataset of the electronic device is first obtained, the damage dataset including accelerated degradation parameters of the electronic device under a preset damaging environment. Secondly, an accelerated degradation model of the electronic device is constructed using a power-law environmental comprehensive stress model and a Wiener degradation model of the electronic device. The accelerated degradation model characterizes the accelerated degradation process of the electronic device under environmental comprehensive stress constructed based on the Wiener process. Based on the accelerated degradation process of the electronic device, an average life prediction model of the electronic device is constructed that integrates the power-law environmental comprehensive stress model and the data-driven model. The data-driven model is generated based on a cumulative distribution function. Furthermore, based on the historical accelerated degradation dataset, the undetermined parameters of the accelerated degradation model are optimized based on a particle swarm algorithm (PSO). The particle swarm algorithm is used to perform a global search within a preset parameter space and determine target estimated values ​​of the undetermined parameters in the accelerated degradation model. Finally, based on the target estimated values ​​of the undetermined parameters in the accelerated degradation model, a damage prediction value of the electronic device under environmental comprehensive stress is determined based on the average life prediction model. The damage prediction value includes an average life prediction value. In this way, an accelerated degradation model and an average life prediction model for electronic equipment under comprehensive environmental stress were constructed. The PSO algorithm was used to solve the target estimated values ​​of the undetermined parameters of the accelerated degradation model. The average life prediction model was further used to determine the average life prediction value, achieving accurate prediction of the electronic equipment life under the influence of multiple factors in a damaging environment. Considering the impact of multiple environmental factors in a damaging environment on life prediction, the accelerated degradation model optimized by the particle swarm algorithm can more accurately consider the combined impact of multiple environmental factors (such as temperature, humidity, fire smoke composition, fire smoke concentration, operating current, etc.) on life, thereby improving the accuracy and reliability of the average life prediction model.

[0049] The embodiments of the present application also provide devices, equipment, and computer-readable storage media corresponding to the above method, which have the same beneficial effects as the above method. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 A flowchart of a method for predicting damage to electronic equipment in a complex post-disaster environment provided by an embodiment of the present application;

[0051] Figure 2 A schematic diagram of a model-related process of a method for predicting damage to electronic equipment in a complex post-disaster environment provided by an embodiment of the present application;

[0052] Figure 3 A flow chart of parameter estimation of a PSO algorithm provided in an embodiment of the present application;

[0053] Figure 4 Schematic diagram of the probability density function (PDF) and cumulative distribution function (CDF) of the predicted average lifespan of electronic devices under 20°C and 95% RH conditions provided in an embodiment of the present application;

[0054] Figure 5 A schematic diagram of the actual degradation path of an electronic device under 20°C and 95% RH conditions provided in an embodiment of the present application;

[0055] Figure 6 A schematic diagram of the structure of a device for predicting damage to electronic equipment in a complex post-disaster environment provided by an embodiment of the present application;

[0056] Figure 7 A schematic diagram of the structure of an electronic equipment damage prediction device in a complex post-disaster environment provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0058] In this application, relational terms such as first and second, etc. are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further limitations, an element defined by the statement "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0059] The current electronic equipment damage prediction methods ignore the complex influence of environmental factors in damaging environments, resulting in low accuracy in electronic equipment life prediction.

[0060] Accurately predicting the lifespan of electronic equipment is not only crucial for assessing the health of electronic equipment after a fire, but also a crucial means of ensuring the continued operation of critical systems after an accident, avoiding secondary disasters, and improving overall rescue effectiveness. It plays an irreplaceable role in improving the safety and reliability of electronic equipment in extreme environments.

[0061] In view of this, an embodiment of the present application provides a method and device for predicting damage of electronic equipment in a complex post-disaster environment. On the basis of the Wiener degradation model of electronic equipment constructed based on the Wiener process, an accelerated degradation model of electronic equipment under environmental comprehensive stress is constructed. Based on the accelerated degradation process of electronic equipment, a power-law environmental comprehensive stress model and an average life prediction model of electronic equipment based on a data-driven model are constructed. According to the historical accelerated degradation data set, the target estimated value of the undetermined parameter of the accelerated degradation model is solved by the PSO algorithm, and the average life prediction model is further used to determine the damage prediction value of the electronic equipment under environmental comprehensive stress, and the damage prediction value includes the average life prediction value. Comprehensively considering various influencing factors in a damaging environment, and fully incorporating environmental influencing factors into life prediction, it can better reflect the degradation process of electronic equipment under actual working conditions, so as to achieve accurate prediction of the electronic equipment life of electronic equipment. Considering the impact of multiple environmental factors on life prediction in a damaging environment, the accelerated degradation model optimized by the particle swarm algorithm can more accurately consider the comprehensive impact of multiple environmental factors (such as temperature, humidity, fire smoke components, fire smoke concentration, working current, etc.) on life, thereby improving the accuracy and reliability of the average life prediction model.

[0062] In order to facilitate understanding of the technical solution provided by the embodiment of the present application, the following describes a method and device for predicting damage to electronic equipment in a complex post-disaster environment provided by the embodiment of the present application in conjunction with the accompanying drawings. Figure 1 , Figure 1 This is a flowchart illustrating a method for predicting damage to electronic equipment in a complex post-disaster environment, as provided in an embodiment of the present application. The method for predicting the lifespan of electronic equipment in a damaging environment can be applied to a server or system, and is not limited in this embodiment of the present application. The prediction method specifically includes steps S101-S105.

[0063] S101: Acquire a damage dataset of an electronic device, where the damage dataset includes accelerated degradation parameters of the electronic device under a preset fire damage environment.

[0064] Fire smoke environment is a type of damaging environment. When electronic equipment is in a post-disaster situation, various factors such as temperature, humidity, and smoke composition will affect the conductive properties of the interconnected structure, thereby affecting the normal operation of the electronic equipment.

[0065] A typical fire smoke environment was simulated in a laboratory setting. By adjusting the temperature, humidity, and smoke composition, test conditions similar to those in actual fire scenarios were established. The changes in the electrical conductivity of electronic devices after a disaster, using a particle swarm optimization algorithm, were used as a historical accelerated degradation dataset. This electrical conductivity can be assessed using the rate of change of on-resistance.

[0066] In some embodiments of the present application, the historical accelerated degradation data set includes accelerated degradation parameters of the electronic device under a preset fire damage environment. The electronic device may include at least one of the following: on-resistance change rate, leakage current, wherein the leakage current refers to the extremely small current generated between the live wire and the neutral wire of the electronic device when the electronic device is operating normally.

[0067] S102: constructing an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and the Wiener degradation model of the electronic device, wherein the accelerated degradation model characterizes an accelerated degradation process of the electronic device under environmental comprehensive stress.

[0068] For example, the Wiener degradation model of the electronic device is constructed based on the Wiener process.

[0069] The Wiener process can be used to describe degradation processes that increase linearly over time and exhibit random fluctuations. It can model non-monotonic degradation data and is applicable to the vast majority of electronic devices that fail due to degradation. In this embodiment, the Wiener process can be used to describe the random degradation process of electronic devices, specifically the degradation behavior of solder joint resistance.

[0070] In one possible implementation, the process of constructing the Wiener degradation model of the electronic device may include:

[0071] Constructing the Wiener degradation model of electronic devices based on the Wiener process:

[0072] Y(t)=Y(0)+μ·τ(t;β)+σ·B(τ(t;β)) (1)

[0073] Where Y(t) is the state of the electronic device at time t, Y(0) is the initial state of the electronic device, and Y(0) = 0 is always true; μ is the drift parameter, which is used to characterize the degradation rate of the electronic device and is closely related to the degradation process of the electronic device, indicating the inter-unit variability caused by the manufacturing or external environment; σ is the diffusion parameter, which characterizes the uncertainty related to time t in the degradation process of the electronic device; B(t) is the standard Brownian motion, which reflects the inherent time-varying randomness in the degradation process; τ(t; β) is the time scale conversion function, τ(t; β) = t β ; β is an undetermined parameter.

[0074] In one possible implementation, when τ(t;β)=t β When the average lifespan of an electronic device is predicted, the probability density function (PDF) f γ (t; μ, σ, D) is:

[0075]

[0076] Wherein, D is a preset failure threshold, indicating that when the degradation parameter of the electronic device reaches D for the first time, the electronic device fails.

[0077] The accelerated degradation model characterizes the accelerated degradation process of the aforementioned electronic device under environmental combined stress, while also considering the impact of ambient temperature and relative humidity on the failure time of the electronic device's conductive performance. In the embodiments of this application, a power-law environmental combined stress model is used to optimize the degradation model, thereby constructing an accelerated degradation model for the electronic device. Specifically, the power-law environmental combined stress model and the Wiener degradation model for the electronic device are used to construct the accelerated degradation model for the electronic device under environmental combined stress.

[0078] The power-law environmental comprehensive stress model can also be called the power-law temperature-humidity model. The power-law environmental comprehensive stress model is mainly used to predict the failure time of materials under different temperature and humidity conditions. The power-law environmental comprehensive stress model can be expressed as:

[0079]

[0080] Among them, t F is the failure time, in h; A0 is the proportionality coefficient related to the material and test conditions; H is the relative humidity, in %; n is an empirical constant used to reflect the degree of influence of humidity on the failure time; E a is the activation energy; R is the Boltzmann constant, R = 8.617 × 10 -5 eV / K; T is the absolute temperature, unit K.

[0081] Specifically, to construct an accelerated degradation model for electronic equipment, it is necessary to first construct an acceleration factor.

[0082] The acceleration factor (AF) is a key parameter in accelerated degradation testing (ADT). It is used to measure the accelerated effect of a specific accelerated environmental stress level in an ADT. It can be defined as the ratio of the lifespan of an electronic device under specific environmental conditions to its lifespan under the accelerated stress level.

[0083] Assume that the electronic device is under stress S k The lifespan under is t k , the cumulative distribution function (CDF) is F k (t k ); at stress S l The lifespan under is t l , CDF is F l (t l). When the CDFs of the two stress levels are equal, that is, F k (t k )=F l (t l ), the kth stress S k Compared with the first stress S l The acceleration factor A k,l for:

[0084]

[0085] According to the Nelson hypothesis, the relationship between the acceleration factor and life can be expressed by CDF or PDF. Taking PDF as an example:

[0086]

[0087] According to f v The expression of (t; μ, σ, D), the acceleration factor A k,l It can be expressed as:

[0088]

[0089] Among them, f k (t k ) is t k The corresponding PDF, f l (t l ) is t l The corresponding PDF.

[0090] According to the principle of constant acceleration factor, that is, the acceleration factor A k,l Only with stress level S k and S l Therefore, the above formula (6) is related to t k The relevant coefficients are all 0:

[0091]

[0092] The acceleration factor A can be derived k,l The relationship between the drift parameter μ and the diffusion parameter σ is:

[0093]

[0094] Since the embodiment of the present application mainly considers the influence of temperature and relative humidity on the degradation process of the conductive performance of electronic devices, a power-law type environmental comprehensive stress model is used to establish the relationship between the parameters of the Wiener degradation model and the environmental comprehensive stress, wherein the environmental comprehensive stress can include temperature stress T and humidity stress H. k and humidity stress H k The drift parameter μ underk and diffusion parameters They are:

[0095]

[0096] Correspondingly, at temperature stress T l and humidity stress H l The drift parameter μl and the diffusion parameter They are:

[0097]

[0098] Where T is the temperature in K; H is the relative humidity in %; a1, a2, b1, b2, c1 and c2 are unknown parameters. Substituting formulas (9)-(12) into formula (8), we can deduce that b1=b2, c1=c2, and the drift parameter μ(T,H) and diffusion parameter σ of the accelerated degradation model are 2 (T,H) is:

[0099]

[0100] According to formula (8), the acceleration factor can be expressed as:

[0101]

[0102] Where, k represents the kth stress level; l represents the lth stress level; μ k is the drift parameter at the kth stress level; μl is the drift parameter at the lth stress level; H k is the kth temperature stress; H l is the lth temperature stress; T k is the kth temperature stress; T l is the lth temperature stress; b1, c1 and β are unknown parameters.

[0103] Based on the above analysis, a power-law environmental comprehensive stress model is used to incorporate the impact of environmental comprehensive stress on the degradation process into the optimization of the degradation model. An accelerated degradation model for electronic devices is constructed, and the drift parameter μ and diffusion parameter σ of the degradation model are associated with the environmental comprehensive stress. Taking humidity stress as an example of non-thermal stress, the Wiener process of environmental temperature stress T and humidity stress H, that is, the accelerated degradation model of electronic devices, can be expressed as:

[0104]

[0105] Where Y(t,S) is the accelerated degradation state of the electronic device under the combined environmental stress at time t, N represents the normal distribution, and a1, a2, b1, c1, and β are unknown parameters.

[0106] S103: Based on the accelerated degradation process of the electronic device, construct an average life prediction model for the electronic device that integrates the power-law environmental comprehensive stress model and the data-driven model.

[0107] The power-law environmental comprehensive stress model and the data-driven model are integrated to construct the average life prediction model of electronic equipment.

[0108] Alternatively, a data-driven model is one that builds a model by learning from data rather than relying on traditional physical principles or assumptions. The Wiener degradation process primarily relies on a damage dataset of the electronic device, from which the data-driven model is constructed.

[0109] The construction process of the average life expectancy prediction model can be seen as follows:

[0110] A failure threshold D is set in advance. When the degradation parameter reaches D for the first time, the electronic device fails. Therefore, the average lifespan of an electronic device can be defined as the time when the degradation process {Y(t), t>0} first reaches the failure threshold, also known as the first arrival time γ. Due to the randomness of the degradation process, the average lifespan γ is a random variable.

[0111] γ=inf{t|Y(t)≥D} (17)

[0112] Define f γ (t) is the PDF of the mean life span γ, F γ (t) is the CDF of the mean lifetime γ. Under the definition of the first arrival time γ, the CDF follows the inverse Gaussian distribution:

[0113]

[0114] in, and is the CDF of the standard normal distribution.

[0115] When τ(t;β)=t β When , the PDF of the first arrival time γ can be found in the above formula (2).

[0116] For continuous random variables, PDF is the derivative of CDF, then the mathematical expectation of the mean life span γ is:

[0117]

[0118] Among them, γ is the average predicted life expectancy, F γ (t; μ, σ, D) is the cumulative distribution function, μ is the drift parameter, σ is the diffusion parameter, and D is the preset failure threshold.

[0119] Further,

[0120]

[0121] Where μ and σ 2 are the target drift parameter and target diffusion parameter of the accelerated degradation model under environmental comprehensive stress. The target drift parameter μ and target diffusion parameter σ of the accelerated degradation model are both related to the environmental stress, and the relationship between the target drift parameter μ and target diffusion parameter σ and the environmental stress is characterized by the power-law environmental comprehensive stress model.

[0122] Use τ = t β Perform variable substitution:

[0123] t=τ 1 / β (twenty one)

[0124]

[0125] By using variable substitution and approximation methods, the mathematical expectation of the first arrival time can be simplified:

[0126]

[0127] S104: Optimizing the undetermined parameters of the accelerated degradation model based on the particle swarm algorithm (PSO) according to the historical accelerated degradation dataset. The particle swarm algorithm is used to perform a global search in a preset parameter space and determine target estimated values ​​of the undetermined parameters in the accelerated degradation model.

[0128] Among them, the particle swarm optimization algorithm can avoid local optimal solutions and determine the target estimated values ​​of the undetermined parameters in the accelerated degradation model by performing a global search in the preset parameter space.

[0129] Specifically, first, based on the expression of the accelerated degradation model of the electronic device, the likelihood function L(a1, a2, b1, c1, β) of the accelerated degradation model is established:

[0130]

[0131] Among them, y ijk t ijk (i.e., the kth stress S k The degradation measurement value of the j-th interconnect structure at the i-th measurement time); Δy ijk =y ijk -y (i-1)jk represents the degradation increment; represents the time increment; i=1,2,…,H jk ; j=1,2,…,N k ; k=1,2,…,M. Hjk The total measurement time of each electronic device under each accelerated environmental comprehensive stress; N k is the total number of electronic devices under each accelerated environmental comprehensive stress; M is the total number of accelerated environmental comprehensive stress.

[0132] In some embodiments of the present application, Figure 3 As shown, step S104 optimizes the undetermined parameters of the accelerated degradation model based on the particle swarm algorithm (PSO) according to the historical accelerated degradation data set. The particle swarm algorithm is used to perform a global search in a preset parameter space and determine the target estimated values ​​of the undetermined parameters in the accelerated degradation model, including:

[0133] A1. Initialize the particle swarm using the particle swarm algorithm (PSO). The initialized particle swarm includes the number of individuals in the particle swarm, the maximum number of iterations, and the search range of each particle. Each particle in the particle swarm represents an undetermined parameter in the accelerated degradation model. The position and velocity of the particle represent the value and change trend of the undetermined parameter, respectively. During initialization, the initial position and velocity of the particle are randomly distributed within the preset parameter space.

[0134] Specifically, the particle swarm is initialized. This process involves setting the number of particles in the swarm, the maximum number of iterations, and the search range for each particle. Each particle represents a potential parameter solution in the accelerated degradation model, with the particle's position and velocity representing the value of the undetermined parameter and its changing trend, respectively. During initialization, the particle's initial position and velocity are randomly distributed within the preset parameter space, providing a diverse parameter solution space for the subsequent optimization process.

[0135] A2. Obtaining a fitness function, which is used to evaluate the degree of closeness between the undetermined parameter represented by each particle and the actual accelerated degradation data;

[0136] Specifically, a fitness function is defined. The fitness function is used to evaluate the quality of the parameter combination represented by each particle. Using a historical accelerated degradation dataset, the fitness function measures the particle's performance based on the model's prediction error. Typically, this process involves calculating the error between the predicted value and the actual accelerated degradation data, with the common goal being to minimize the error or maximize the likelihood function value. Specifically, the fitness function is typically calculated as the sum of squared errors, expressed as the deviation between the predicted degradation value and the actual data.

[0137] A3. Search and update the speed and position of each particle in the particle swarm to iteratively approach the global optimal solution. In each iteration, the speed and position of the particle are updated according to the particle update rule to guide the particle swarm to move towards the global optimal position. In each update, the speed and position of the particle are affected by the current particle's own historical best position, the global best position, and random factors, ensuring that the particle swarm performs an effective global search in the search space.

[0138] Specifically, the particle swarm enters the search and update phase. The core of the particle swarm algorithm lies in updating the particle's position and velocity, gradually approaching the global optimal solution through iteration. During each iteration, the particle's velocity and position are updated according to specific rules to guide the particle swarm towards the global optimal position. During each update, the particle's velocity and position are influenced by the particle's own historical best position, the global best position, and certain random factors, ensuring that the particle swarm conducts an effective global search within the search space.

[0139] A4. After each iteration, the fitness of each particle is evaluated using a fitness function by using the historical accelerated degradation data set and the prediction error of the accelerated degradation model to determine whether the undetermined parameters of the accelerated degradation model corresponding to the current position of each particle can effectively fit the historical accelerated degradation data;

[0140] Specifically, after each iteration, the fitness of each particle is evaluated. By calculating the fitness function, we can determine whether the accelerated degradation model parameters corresponding to each particle's current position can effectively fit the historical accelerated degradation data. If a particle's position minimizes the fitness function or maximizes the likelihood function, it indicates that the parameter combination corresponding to that particle is close to the true value.

[0141] As iterations progress, the particle swarm continuously updates each particle's historical best position and the global best position. In each iteration, particles adjust their positions based on their fitness evaluation results, gradually converging toward the optimal solution. Through multiple iterations, the particle swarm discovers an optimal parameter combination that makes the accelerated degradation model's predictions as close as possible to historical data, thereby obtaining the optimal undetermined parameters.

[0142] A5. After reaching a predetermined maximum number of iterations or satisfying a convergence condition, outputting target estimated values ​​of undetermined parameters of the optimized accelerated degradation model.

[0143] Finally, after reaching a predetermined maximum number of iterations or satisfying convergence criteria, the algorithm outputs the optimized accelerated degradation model parameters. These parameters are used in subsequent steps to determine the acceleration factor of electronic devices under different environmental conditions and ultimately calculate the average lifespan prediction of electronic devices. Through optimization using the particle swarm algorithm, the prediction accuracy and reliability of the accelerated degradation model can be significantly improved, providing more accurate results for electronic device lifespan prediction.

[0144] Specifically, in order to obtain the target estimated values ​​of the undetermined parameters, a particle swarm optimization (PSO) algorithm is used to estimate the target estimated values ​​of the undetermined parameters.

[0145] Assume that the particle swarm contains N particles, each particle represents a possible parameter combination (a1, a2, b1, c1, β), that is, the position of the particle is (a1, a2, b1, c1, β). Then initialize the initial position and velocity of the particle within the appropriate parameter space.

[0146] Then define the fitness function of the particle swarm, which is the objective function of the maximum likelihood estimation function. In this case, the fitness function is the logarithm of the likelihood function.

[0147] The core of the particle swarm algorithm is to search for the optimal solution by continuously updating the position and velocity of particles. The update formula for each particle is as follows:

[0148] Update speed formula:

[0149]

[0150] Update speed formula:

[0151]

[0152] in, and are the velocity and position of particle i at step n respectively; is the historical best position of particle i at step n; g n is the global optimal position of the current particle swarm; w is the inertia weight, which is usually used to control the exploration ability of the particles; m1 and m2 are acceleration constants, which control the speed at which the particles move to their own optimal position and the global optimal position; r1 and r2 are random numbers between 0 and 1.

[0153] Through the iterative process of the particle swarm algorithm, the global optimal solution g is continuously updated. n , and eventually the particle swarm will converge to the global optimal solution, which is the parameter combination that maximizes the likelihood function This is the target estimate of the unknown parameter.

[0154] S105: Determine a damage prediction value of the electronic device under environmental comprehensive stress based on the average life prediction model according to target estimated values ​​of undetermined parameters in the accelerated degradation model, wherein the damage prediction value includes an average life prediction value.

[0155] Through step S104, the target estimated value of the undetermined parameter in the accelerated degradation model can be obtained, and the acceleration factor under the environmental comprehensive stress can be obtained based on the target estimated value of the undetermined parameter. Determine the target coefficient and obtain the acceleration factor under the environmental comprehensive stress according to the target coefficient. The target coefficient includes the drift parameter μ(T,H) and the diffusion parameter σ 2 (T,H).

[0156] The target estimated value of the undetermined parameter in this accelerated degradation model can be used as an acceleration factor, which is the ratio of the life characteristic value of the electronic device under environmental comprehensive stress to the life characteristic value under normal stress. By introducing the acceleration factor, the simulated failure process of the electronic device is accelerated.

[0157] The average life prediction model can predict the damage prediction value of the electronic device under environmental comprehensive stress based on the acceleration factor. The damage prediction value can at least include the average life prediction value. Without limitation, the damage prediction value of the electronic device can also include: a prediction value of the degree of surface damage of the electronic device, for example, a prediction value of cracks on the surface of the electronic device.

[0158] Average life prediction value. Through the acceleration factor, the failure time under different environmental comprehensive stress levels can be predicted.

[0159] As mentioned above, the damage prediction method for electronic equipment in a damaging environment includes the construction of a degradation model, an accelerated degradation model, and an average life prediction model. Figure 2 , Figure 2 A schematic diagram of the model-related flow of a method for predicting damage to electronic equipment in a complex post-disaster environment provided in an embodiment of the present application.

[0160] Taking the nonlinear Wiener process as an example, the degradation law can be analyzed based on the degradation process of electronic equipment. According to the degradation law of the ball grid array (BGA) package interconnect structure and the nonlinear Wiener process, a Wiener degradation model of electronic equipment can be constructed.

[0161] On the basis of the degradation model, the acceleration factor is defined, the relationship between the acceleration factor and the degradation model parameters is calculated, and then the relationship between the degradation model parameters and environmental factors is constructed to establish the accelerated degradation model.

[0162] The PDF and CDF of the average lifespan, i.e., the average lifespan prediction model, are established. The PSO algorithm and the accelerated degradation dataset are applied to determine the unknown parameters to calculate the average lifespan prediction value of the electronic device.

[0163] The specific implementation of the above model construction and use can be found in the description of the above embodiments and will not be repeated here.

[0164] Based on the contents of steps S101-S105 above, it can be seen that based on the changes in the conductivity of electronic devices in damaging environments, multiple influencing factors such as temperature, humidity, and smoke composition in damaging environments are comprehensively considered. Incorporating these factors into the construction and use of the degradation model and average life prediction model can better reflect the degradation process of electronic devices under actual operating conditions and accurately predict the lifespan of electronic devices in different damaging environments through numerical calculations of the model. In this way, the consideration of multiple factors makes the model used in the embodiments of the present application more applicable and can effectively operate under different fire environmental conditions. It can also combine the changes in the conductivity of electronic devices detected in real time to achieve high-precision lifespan prediction, reducing errors caused by environmental complexity and variability. It provides reliable data support for the continued use of electronic devices after a fire, avoiding the additional risks and costs caused by premature or delayed equipment replacement. Considering the impact of multiple environmental factors in damaging environments on lifespan prediction, the accelerated degradation model optimized by the particle swarm algorithm can more accurately consider the combined impact of multiple environmental factors (such as temperature and humidity) on lifespan, thereby improving the accuracy and reliability of the average lifespan prediction model.

[0165] The embodiments of this application can also be integrated with existing lifespan assessment systems to implement real-time health monitoring and lifespan warnings for electronic equipment in fire environments. In high-risk fire scenarios, applying the solutions provided by the embodiments of this application can improve the reliability of electronic equipment, predict possible faults and failures in advance, and take timely countermeasures to avoid secondary disasters.

[0166] It should be noted that the embodiments of the present application are not only applicable to damaging environments, but can also be extended to the life prediction of electronic equipment in other harsh environments, which can improve the safety and reliability of the equipment and further enhance the level of full life cycle management of the equipment.

[0167] Taking electronic equipment as an example, the following describes a method for predicting damage to electronic equipment in a complex post-disaster environment provided by an embodiment of the present application through a specific experimental process.

[0168] The degradation law of the conductive performance of electronic equipment is selected as the accelerated degradation data set, and the particle swarm optimization (PSO) method is used to estimate the target estimated values ​​of the undetermined parameters of the accelerated degradation model of electronic equipment.

[0169]

[0170] For a description of the particle swarm PSO method, see the previous Figure 3 And corresponding examples.

[0171] According to the relationship between the target estimated value of the undetermined parameter and the coupling stress, the target coefficients μ(T,H) and σ under the specific environmental conditions of 20℃ and 95% relative humidity (RH) can be obtained. 2 (T, H). Substituting the target coefficient into the average life prediction model can obtain the corresponding average life prediction value of the electronic device, as shown in Table 1 below.

[0172] parameter 20℃-95%RH μ 0.0061 <![CDATA[σ 2 ]]> 0.0013 Average life expectancy 84.52

[0173] Table 1 Target coefficients and corresponding average life expectancy prediction values

[0174] The average life expectancy of electronic equipment under the conditions of 20°C and 95% RH is predicted to be 84.52 hours (h).

[0175] At the same time, you can also get the PDF and CDF of the average life expectancy prediction value, see Figure 4 , Figure 4 The PDF and CDF curves in the figure show that the average life expectancy of electronic equipment is concentrated in the range of 70-110 hours, which is a high-risk area for electronic equipment failure. Figure 5 As shown in the figure, three sets of degradation data were measured at 20°C and 95% RH, and the corresponding experimental values ​​for the average lifespan of the electronic device were obtained. The degradation path first reached the failure threshold at 85.07 hours (h), indicating that the actual average lifespan of the electronic device is 85.07 hours.

[0176] The above data shows that the relative error between the average lifespan prediction model and the actual measurement results is 0.64%, which is below 10%, which is relatively low. Therefore, the average lifespan prediction model based on the random Wiener process in the embodiment of the present application is more accurate and applicable under a wide range of ambient temperature and relative humidity conditions.

[0177] Accurately predicting the lifespan of electronic equipment is the key to assessing the health status of electronic equipment after a fire. It can also ensure the continuous operation of key systems after an accident, avoid secondary disasters, and improve the overall rescue effect. The technical solution of the embodiment of the present application can not only provide strong support for the reliability assessment of electronic equipment in a damaging environment, but also provide theoretical and technical references for research and engineering applications in related fields. Taking into account the impact of various environmental factors on life prediction in a damaging environment, the accelerated degradation model optimized by the particle swarm algorithm can more accurately consider the comprehensive impact of various environmental factors (such as temperature, humidity, fire smoke components, fire smoke concentration, working current, etc.) on life, thereby improving the accuracy and reliability of the average life prediction model.

[0178] The above embodiment of the present application provides a method for predicting damage of electronic equipment in a complex post-disaster environment. Next, the present application also provides a device for predicting damage of electronic equipment in a complex post-disaster environment, which is used to perform the above Figure 1 The function of the electronic equipment damage prediction device in a complex post-disaster environment under a damaging environment is described. The structural diagram of the device is shown in FIG. Figure 6 As shown, it includes an acquisition module 601, a construction module 602, a solution module 603 and a prediction module 604.

[0179] in,

[0180] An acquisition module 601 is configured to acquire a damage dataset of an electronic device, wherein the damage dataset includes accelerated degradation parameters of the electronic device under a preset damaging environment.

[0181] A construction module 602 is configured to construct an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and a Wiener degradation model of the electronic device, wherein the accelerated degradation model characterizes an accelerated degradation process of the electronic device under environmental comprehensive stress;

[0182] The construction module 602 is further configured to construct an average life prediction model for electronic devices based on the accelerated degradation process of the electronic devices, integrating the power-law environmental comprehensive stress model and the data-driven model;

[0183] A solution module 603 is configured to optimize the undetermined parameters of the accelerated degradation model based on the historical accelerated degradation data set using a particle swarm algorithm (PSO). The particle swarm algorithm is configured to determine target estimated values ​​of the undetermined parameters in the accelerated degradation model by performing a global search within a preset parameter space.

[0184] The prediction module 604 is used to determine the damage prediction value of the electronic device under environmental comprehensive stress based on the target estimated value of the undetermined parameter in the accelerated degradation model and the average life prediction model, wherein the damage prediction value includes the average life prediction value.

[0185] In one possible implementation, the construction module 602 is configured to construct an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and the Wiener degradation model of the electronic device, including:

[0186] A submodule is constructed to simulate the degradation path of the electronic device using the Wiener degradation model of the electronic device, and to associate the drift parameter μ and diffusion parameter σ of the Wiener degradation model with the environmental comprehensive stress using the power-law environmental comprehensive stress model to construct an accelerated degradation model of the electronic device.

[0187] The drift parameter μ(T,Q i ) is expressed as Diffusion parameter σ 2 (T,Q i ) is expressed as The accelerated degradation model of the electronic device is: The environmental comprehensive stress includes temperature stress T and non-thermal stress Q, wherein the non-thermal stress can be one or more of humidity stress, current stress, mechanical stress and electromagnetic stress, and n represents the number of non-thermal stresses; Y(t, S) is the accelerated degradation state of the electronic device under the action of environmental comprehensive stress at time t, N represents the normal distribution, a1, a2, b i , c1 and β are unknown parameters.

[0188] In one possible implementation, the solution module 603 includes:

[0189] An initialization submodule is used to initialize a particle swarm according to the particle swarm algorithm (PSO). The initialized particle swarm includes: the number of individuals in the particle swarm, the maximum number of iterations, and the search range of each particle. Each particle in the particle swarm represents an undetermined parameter in the accelerated degradation model. The position and velocity of the particle represent the value and change trend of the undetermined parameter, respectively. During initialization, the initial position and velocity of the particle are randomly distributed within a preset parameter space.

[0190] An acquisition submodule, used for acquiring a fitness function, wherein the fitness function is used to evaluate the degree of proximity between the undetermined parameter represented by each particle and the actual accelerated degradation data;

[0191] A search and update submodule is used to search and update the speed and position of each particle in the particle swarm to iteratively approach the global optimal solution. In each iteration, the speed and position of the particle are updated according to the particle update rule to guide the particle swarm to move towards the global optimal position. In each update, the speed and position of the particle are affected by the current particle's own historical best position, the global best position, and random factors, ensuring that the particle swarm performs an effective global search in the search space.

[0192] an evaluation submodule, configured to evaluate the fitness of each particle after each iteration by using the historical accelerated degradation data set and the prediction error of the accelerated degradation model using the fitness function to determine whether the undetermined parameters of the accelerated degradation model corresponding to the current position of each particle can effectively fit the historical accelerated degradation data;

[0193] The output submodule is used to output the target estimated values ​​of the undetermined parameters of the optimized accelerated degradation model after reaching a predetermined maximum number of iterations or satisfying a convergence condition.

[0194] In one possible implementation, the construction module 602 is further configured to construct, based on the accelerated degradation process of the electronic device, an average life prediction model for the electronic device that integrates the power-law environmental comprehensive stress model and a data-driven model, wherein the data-driven model is generated based on a cumulative distribution function and includes:

[0195] A fusion submodule is used to fuse the power-law environmental comprehensive stress model and the data-driven model to construct an average life prediction model for the electronic device. The average life prediction model is: Among them, γ is the average predicted life expectancy, F γ (t; μ, σ, D) is the cumulative distribution function at time t, μ is the target drift parameter of the accelerated degradation model, σ 2 is the target diffusion parameter of the accelerated degradation model, and D is the preset failure threshold.

[0196] It should be noted that the steps executed by each module in the electronic equipment damage prediction device in a complex post-disaster environment provided in an embodiment of the present application and the related technical features correspond to the method provided in the embodiment of the application. The description of the device part can be found in the embodiment of the aforementioned method part and will not be repeated here.

[0197] An embodiment of the present application provides a device for predicting damage to electronic devices in a complex post-disaster environment, the device comprising an acquisition module, a construction module, a solution module, and a prediction module. The acquisition module is used to acquire a historical accelerated degradation dataset of the electronic device, wherein the historical accelerated degradation dataset comprises accelerated degradation parameters of the electronic device under a preset damaging environment. The construction module is used to construct an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and a Wiener degradation model of the electronic device, wherein the accelerated degradation model characterizes the accelerated degradation process of the electronic device under environmental comprehensive stress. The construction module is also used to construct an average life prediction model of the electronic device based on a power-law environmental comprehensive stress model and a data-driven model based on the accelerated degradation process of the electronic device, wherein the data-driven model is generated based on a cumulative distribution function. The solution module is used to optimize the undetermined parameters of the accelerated degradation model based on the historical accelerated degradation dataset and a particle swarm algorithm (PSO), wherein the particle swarm algorithm is used to determine the target estimated values ​​of the undetermined parameters in the accelerated degradation model by performing a global search in a preset parameter space. The prediction module is used to determine the damage prediction value of the electronic device under environmental comprehensive stress based on the average life prediction model according to the target estimated values ​​of the undetermined parameters in the accelerated degradation model, and the damage prediction value includes the average life prediction value. In this way, an accelerated degradation model and an average life prediction model of the electronic device under environmental comprehensive stress are constructed, and the target estimated values ​​of the undetermined parameters of the accelerated degradation model are solved by the PSO algorithm. The average life prediction value is further determined by the average life prediction model, thereby achieving an accurate prediction of the electronic device life under the influence of multiple factors in a damaging environment. Considering the impact of various environmental factors on life prediction in a damaging environment, the accelerated degradation model optimized by the particle swarm algorithm can more accurately consider the comprehensive impact of various environmental factors (such as temperature, humidity, fire smoke components, fire smoke concentration, working current, etc.) on life, thereby improving the accuracy and reliability of the average life prediction model.

[0198] Based on the above method embodiment, a method for predicting damage to electronic equipment in a complex post-disaster environment is provided. The present application embodiment provides a device for predicting damage to electronic equipment in a complex post-disaster environment. Figure 7 , the device includes: a processor, a memory, and a system bus;

[0199] The processor and the memory are connected via the system bus;

[0200] The memory is used to store one or more programs, and the one or more programs include instructions. When the instructions are executed by the processor, the processor executes the method for predicting damage of electronic equipment in a complex post-disaster environment as described in any one of the above embodiments.

[0201] Based on a method for predicting damage of electronic equipment in a complex post-disaster environment provided by the above-mentioned method embodiment, an embodiment of the present application provides a computer-readable storage medium, which stores instructions. When the instructions are executed on a device, the device executes the method for predicting damage of electronic equipment in a complex post-disaster environment described in any of the above-mentioned embodiments.

[0202] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Similar parts between the various embodiments can be referenced to each other.

[0203] Those skilled in the art will understand that Figure 1 and Figure 3 The flowchart shown is only an example in which the embodiments of the present application can be implemented, and the scope of application of the embodiments of the present application is not limited in any aspect by the flowchart.

[0204] In the several embodiments provided in this application, it should be understood that the disclosed methods, devices and equipment can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another device or system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of devices or units, which can be electrical connection, mechanical connection or other forms.

[0205] The units described as separate components may or may not be physically separate, and the components displayed 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 may be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0206] If the functions are implemented in the form of 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 the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0207] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting damage to electronic equipment in a complex post-disaster environment, characterized in that: The damage prediction method comprises: Acquiring a damage dataset of an electronic device, the damage dataset including accelerated degradation parameters of the electronic device under a preset fire damage environment; Constructing an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and the Wiener degradation model of the electronic device, wherein the accelerated degradation model characterizes the accelerated degradation process of the electronic device under the environmental comprehensive stress; Based on the accelerated degradation process of the electronic device, constructing an average life prediction model for the electronic device that integrates the power-law environmental comprehensive stress model and the data-driven model; According to the historical accelerated degradation data set, the undetermined parameters of the accelerated degradation model are optimized based on the particle swarm algorithm (PSO). The particle swarm algorithm is used to perform a global search in a preset parameter space and determine the target estimated values ​​of the undetermined parameters in the accelerated degradation model; Determining a damage prediction value of the electronic device under environmental comprehensive stress based on the average life prediction model according to target estimated values ​​of undetermined parameters in the accelerated degradation model, wherein the damage prediction value includes an average life prediction value; The method of constructing an average life prediction model for electronic devices based on the accelerated degradation process of the electronic devices, which integrates the power-law environmental comprehensive stress model and the data-driven model, includes: The power-law environmental comprehensive stress model and the data-driven model are integrated to construct the average life prediction model of the electronic device. The average life prediction model is: ， in, is the predicted life expectancy, for The cumulative distribution function of time, To accelerate the target drift parameter of the degradation model, is the target diffusion parameter of the accelerated degradation model, is the preset failure threshold, represents a nonlinear function of time t, Used to describe the nonlinear characteristics of electronic equipment degradation over time; the target drift parameter of the accelerated degradation model and target diffusion parameters are all related to environmental stress, the target drift parameters and target diffusion parameters The relationship with environmental stress is characterized by the power-law environmental comprehensive stress model.

2. The damage prediction method according to claim 1, characterized in that: The method of constructing the accelerated degradation model of the electronic device by using the power-law environmental comprehensive stress model and the Wiener degradation model of the electronic device includes: The Wiener degradation model of the electronic device is used to simulate the degradation path of the electronic device, and the drift parameter of the Wiener degradation model is used to simulate the degradation path of the electronic device. and diffusion parameters In association with the comprehensive environmental stress, an accelerated degradation model of the electronic device is constructed; wherein the comprehensive environmental stress includes temperature stress, humidity stress, current stress, mechanical stress, and electromagnetic stress.

3. The damage prediction method according to claim 2, characterized in that: The method optimizes the undetermined parameters of the accelerated degradation model based on the particle swarm algorithm (PSO) according to the historical accelerated degradation data set. The particle swarm algorithm is used to perform a global search in a preset parameter space and determine the target estimated values ​​of the undetermined parameters in the accelerated degradation model, including: Initializing a particle swarm according to the particle swarm algorithm (PSO), wherein the initialized particle swarm includes: the number of individuals in the particle swarm, the maximum number of iterations, and the search range of each particle, wherein each particle in the particle swarm represents an undetermined parameter in the accelerated degradation model, and the position and velocity of the particle respectively represent the value and change trend of the undetermined parameter. During initialization, the initial position and velocity of the particle are randomly distributed within a preset parameter space; Obtaining a fitness function, wherein the fitness function is used to evaluate the degree of proximity between the undetermined parameter represented by each particle and the actual accelerated degradation data; Searching and updating the speed and position of each particle in the particle swarm to iteratively approach the global optimal solution. In each iteration, the speed and position of the particle are updated according to the particle update rule to guide the particle swarm to move towards the global optimal position. In each update, the speed and position of the particle are affected by the current particle's own historical best position, the global best position, and random factors, ensuring that the particle swarm performs an effective global search in the search space. After each iteration, the fitness of each particle is evaluated using the fitness function by using the historical accelerated degradation data set and the prediction error of the accelerated degradation model to determine whether the undetermined parameters of the accelerated degradation model corresponding to the current position of each particle can effectively fit the historical accelerated degradation data; After reaching a predetermined maximum number of iterations or satisfying a convergence condition, target estimated values ​​of undetermined parameters of the optimized accelerated degradation model are output.

4. A device for predicting damage to electronic equipment in a complex post-disaster environment, characterized in that: The electronic equipment damage prediction device comprises: an acquisition module, configured to acquire a damage dataset of an electronic device, wherein the damage dataset includes accelerated degradation parameters of the electronic device under a preset damaging environment; A construction module is used to construct an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and a Wiener degradation model of the electronic device, wherein the accelerated degradation model characterizes the accelerated degradation process of the electronic device under the environmental comprehensive stress; The construction module is further used to construct an average life prediction model for electronic devices that integrates the power-law type environmental comprehensive stress model and the data-driven model based on the accelerated degradation process of the electronic devices; A solution module, configured to optimize the undetermined parameters of the accelerated degradation model based on a particle swarm algorithm (PSO) according to a historical accelerated degradation data set. The particle swarm algorithm is configured to determine target estimated values ​​of the undetermined parameters in the accelerated degradation model by performing a global search within a preset parameter space. A prediction module for determining a damage prediction value of the electronic device under environmental comprehensive stress based on the average life prediction model according to target estimated values ​​of undetermined parameters in the accelerated degradation model, the damage prediction value including the average life prediction value; The building blocks include: A fusion submodule is used to fuse the power-law environmental comprehensive stress model and the data-driven model to construct the average life prediction model of the electronic device, where the average life prediction model is: ， in, is the predicted life expectancy, for The cumulative distribution function of time, To accelerate the target drift parameter of the degradation model, is the target diffusion parameter of the accelerated degradation model, is the preset failure threshold, represents a nonlinear function of time t, Used to describe the nonlinear characteristics of electronic equipment degradation over time; the target drift parameter of the accelerated degradation model and target diffusion parameters are all related to environmental stress, the target drift parameters and target diffusion parameters The relationship with environmental stress is characterized by the power-law environmental comprehensive stress model.

5. The electronic equipment damage prediction device according to claim 4, characterized in that: The construction module is used to construct an accelerated degradation model of the electronic device using a power-law environmental comprehensive stress model and the Wiener degradation model of the electronic device, including: A submodule is constructed to simulate the degradation path of the electronic device using the Wiener degradation model of the electronic device, and to convert the drift parameters of the Wiener degradation model into and diffusion parameters In association with the comprehensive environmental stress, an accelerated degradation model of the electronic device is constructed; wherein the comprehensive environmental stress includes temperature stress, humidity stress, current stress, mechanical stress, and electromagnetic stress.

6. The electronic equipment damage prediction device according to claim 5, characterized in that: The solution module includes: An initialization submodule is used to initialize a particle swarm according to the particle swarm algorithm (PSO). The initialized particle swarm includes: the number of individuals in the particle swarm, the maximum number of iterations, and the search range of each particle. Each particle in the particle swarm represents an undetermined parameter in the accelerated degradation model. The position and velocity of the particle represent the value and change trend of the undetermined parameter, respectively. During initialization, the initial position and velocity of the particle are randomly distributed within a preset parameter space. An acquisition submodule, used for acquiring a fitness function, wherein the fitness function is used to evaluate the degree of proximity between the undetermined parameter represented by each particle and the actual accelerated degradation data; A search and update submodule is used to search and update the speed and position of each particle in the particle swarm to iteratively approach the global optimal solution. In each iteration, the speed and position of the particle are updated according to the particle update rule to guide the particle swarm to move towards the global optimal position. In each update, the speed and position of the particle are affected by the current particle's own historical best position, the global best position, and a random factor, ensuring that the particle swarm performs an effective global search in the search space. an evaluation submodule, configured to evaluate the fitness of each particle after each iteration by using the historical accelerated degradation data set and the prediction error of the accelerated degradation model using the fitness function to determine whether the undetermined parameters of the accelerated degradation model corresponding to the current position of each particle can effectively fit the historical accelerated degradation data; The output submodule is used to output the target estimated values ​​of the undetermined parameters of the optimized accelerated degradation model after reaching a predetermined maximum number of iterations or satisfying a convergence condition.

7. A device for predicting damage to electronic equipment in a complex post-disaster environment, characterized in that: The device includes: a processor, a memory, and a system bus; The processor and the memory are connected via the system bus; The memory is used to store one or more programs, and the one or more programs include instructions. When the instructions are executed by the processor, the processor executes the method for predicting damage of electronic equipment in a complex post-disaster environment according to any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores instructions, and when the instructions are executed on a device, the device executes the method for predicting damage of electronic equipment in a complex post-disaster environment according to any one of claims 1 to 3.

Images