Servo controller accelerated degradation test method based on multi-parameter optimization

Abstract

The invention discloses a servo controller accelerated degradation test method based on multi-parameter optimization. The method comprises the following steps: constructing a servo controller multi-parameter accelerated degradation model: constructing the servo controller multi-parameter accelerated degradation model through a Wiener process model, an Arrhenius model and a Gaussian Copula function; constructing a multi-parameter comprehensive optimal criterion; deducing a service life distribution function of the servo controller at a normal working temperature; on the basis of a V-optimal criterion, a multi-parameter comprehensive optimal criterion is constructed by taking minimization of the quantile life asymptotic variance of the servo controller as an optimization target so as to form an optimization target function; and constructing and solving an accelerated degradation test optimization model: constructing the accelerated degradation test optimization model by taking the optimization target function as a target, and solving the accelerated degradation test optimization model by adopting a genetic algorithm to obtain a servo controller accelerated degradation test scheme. The evaluation precision of predicting the service life and reliability of the servo controller based on the test data is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of accelerated degradation testing technology, and specifically to an accelerated degradation testing method for servo controllers based on multi-parameter optimization. Background Technology

[0002] As a core control unit in industrial automation, hydraulic vibration control, and other fields, the performance stability and reliability of servo controllers directly determine the operating accuracy, dynamic response speed, and service life of downstream equipment. In actual use, ambient temperature accelerates component aging, induces parameter drift, and causes thermal stress damage, leading to performance degradation of core components such as power devices, control chips, and sensors. This, in turn, causes a continuous deterioration of core performance parameters of servo control, such as dynamic response capability and proportional regulation characteristics, affecting the normal operation of equipment. However, the performance degradation process of servo controllers under natural environments is lengthy, making it difficult to quickly assess the lifespan and reliability of servo controllers. There is an urgent need to conduct accelerated degradation tests under temperature stress to shorten the testing cycle and provide a scientific basis for product reliability design and lifespan prediction.

[0003] Servo controllers typically possess numerous performance parameters, and deviations in any of these parameters can lead to malfunctions. Among these, the servo control frequency response time and P-value (proportional gain) are key parameters reflecting its core control capabilities. The former determines the servo controller's dynamic tracking speed and signal transmission delay characteristics; an increase in its value leads to system dynamic response lag and increased overshoot, affecting the device's real-time control accuracy. The latter's stability directly relates to the system's steady-state error and response sensitivity; a drift in its value can cause control algorithm mismatch, resulting in increased positioning errors and aggravated torque fluctuations. Furthermore, a drift in the P-value directly alters the system's gain characteristics, causing oscillating overshoot or sluggishness in the frequency response time. A degradation in the frequency response time reduces the system's perception and feedback speed to signal changes, making it impossible to adjust the P-value in a timely manner to match dynamic operating conditions, further amplifying steady-state errors and dynamic tracking deviations. Therefore, it is necessary to comprehensively consider the requirements of both the servo controller's frequency response time and P-value performance parameters, taking into account the coupling effects between parameters, and conduct accelerated degradation testing and optimization design for the servo controller.

[0004] However, previous accelerated degradation test design methods were mainly aimed at products with a single performance parameter, with the optimization goal of minimizing the variance of model parameter estimation or the error of life index assessment. This not only makes it difficult to meet the multi-performance parameter assessment requirements of servo controllers, but also fails to characterize the impact of the coupling effect between parameters on the test design. As a result, it is difficult to guide the optimization design of accelerated degradation tests for servo controllers, which affects the accurate assessment of their life and reliability. Summary of the Invention

[0005] This invention addresses the reliability assessment requirements of servo controllers under multiple performance parameter requirements by providing a method for accelerated degradation testing of servo controllers based on multi-parameter optimization.

[0006] Terminology Explanation:

[0007] 1. Wiener process model: The Wiener process, also known as Brownian motion, is an important continuous-time stochastic process that is widely used in finance, physics, biology and other fields.

[0008] 2. V-optimal criterion: An experimental design method that aims to minimize the average of the predicted variance of the regression equation, thereby improving the prediction accuracy of the model.

[0009] 3. Gaussian Copula function: This is a statistical tool used to describe the dependencies between multiple random variables. It captures the nonlinear correlations between variables by decomposing the joint distribution of random variables into marginal distributions and dependency structures (Copula functions).

[0010] 4. Fisher Information Matrix (FIM): A crucial tool in statistics and information theory for measuring the amount of information observed data provides about unknown parameters. It characterizes the sensitivity of a probability distribution to changes in parameters and has wide applications in parameter estimation, maximum likelihood estimation (MLE), the Cramér-Rao lower bound, and machine learning optimization.

[0011] The technical solution adopted by this invention to overcome its technical problems is:

[0012] A method for accelerating the degradation of a servo controller based on multi-parameter optimization, comprising the following steps:

[0013] Constructing a multi-parameter accelerated degradation model for a servo controller: Based on the Wiener process model, a degradation process model for each performance parameter of the servo controller is constructed; the influence of temperature stress on each performance parameter is characterized using the Arrhenius model; and the coupling relationship between each performance parameter is characterized using the Gaussian Copula function, thus obtaining a multi-parameter accelerated degradation model for the servo controller.

[0014] Constructing a multi-parameter comprehensive optimal criterion: Based on the multi-parameter accelerated degradation model of the servo controller, derive the lifetime distribution function of the servo controller at normal operating temperature; based on the V-optimal criterion, construct a multi-parameter comprehensive optimal criterion to form the optimization objective function with minimizing the asymptotic variance of the servo controller's quantile lifetime as the optimization objective;

[0015] Constructing and solving an accelerated degradation test optimization model: Determine the test design variables, and under the test cost constraint, construct an accelerated degradation test optimization model with the aforementioned optimization objective function as the objective. Use a genetic algorithm to solve the accelerated degradation test optimization model to obtain the servo controller accelerated degradation test scheme.

[0016] Preferably, the performance parameters of the servo controller include at least frequency response time and P-value. The performance parameters mainly studied in this invention are frequency response time and P-value.

[0017] Preferably, the construction of the degradation process model for each performance parameter of the servo controller based on the Wiener process model specifically includes:

[0018] Based on the Wiener process model, a degradation process model of the servo controller is constructed, including two performance parameters: frequency response time and P-value. The expression is as follows:

[0019] (1)

[0020] In equation (1), Representing the One performance parameter in The level of degradation at any given moment , Representing the The initial values ​​of each performance parameter, and Representing the first The drift coefficient and diffusion coefficient of the performance parameter degradation process model. Represents standard Brownian motion;

[0021] From the properties of the Wiener process, we know that the first... One performance parameter in Degradation increment at time It follows a normal distribution, that is: The probability density function and cumulative distribution function distribution As shown in equations (2) and (3) respectively:

[0022] (2)

[0023] (3)

[0024] In equation (3), The cumulative distribution function represents the standard normal distribution.

[0025] Furthermore, the description of the influence of temperature stress on various performance parameters using the Arrhenius model specifically includes:

[0026] The Arrhenius model was used to characterize the effects of temperature stress on two performance parameters: frequency response time and P-value. The drift coefficient of the degradation process model was then used as the performance parameter. As a parameter of temperature effect, the constructed acceleration model function is shown in equation (4):

[0027] (4)

[0028] In equation (4), Represents absolute temperature. Indicates temperature Next The drift coefficient values ​​of the performance parameter degradation process model. and They respectively represent the corresponding to the first Two parameters of the acceleration model with one performance parameter;

[0029] To avoid the impact of dimensional differences on the experimental optimization calculations, the temperature stress is standardized as shown in equation (5):

[0030] (5)

[0031] In equation (5), This represents the standardized temperature level value. and These represent the normal operating temperature level and the extreme temperature level of the servo controller, respectively.

[0032] Therefore, equation (4) can be further expressed as:

[0033] (6).

[0034] Preferably, the step of using Gaussian Copula functions to characterize the coupling relationship between various performance parameters to obtain a multi-parameter accelerated degradation model of the servo controller specifically includes:

[0035] By using the binary Gaussian Copula function to characterize the coupling relationship between the two performance parameters of frequency response time and P-value, the multi-parameter accelerated degradation model of the servo controller is obtained as shown in equation (7):

[0036] (7)

[0037] In equation (7), Represents the correlation coefficient. , , This represents the inverse function of the cumulative distribution function of the standard normal distribution. The probability density function represents the joint distribution of the degradation increments of the two performance parameters, frequency response time and P-value.

[0038] Preferably, the derivation of the lifetime distribution function of the servo controller at normal operating temperature based on the multi-parameter accelerated degradation model of the servo controller specifically includes:

[0039] set up Represents the number of test samples for the servo controller. Represents the temperature stress level number. Representing the A temperature stress level value, Representing the Number of test samples under each temperature stress level Representing the The number of times the performance parameter values ​​are measured under each temperature stress level, and the measurement interval is [number]. ,but Representing the Under the first temperature stress level, the first The sample at the th The first measurement One performance parameter value, Representing the Under the first temperature stress level The drift coefficient of the i-th performance parameter value, the th Temperature stress level After standardization ,in, , , , ;

[0040] Given that the failure of either the frequency response time or the P-value in the servo controller is considered a system failure, let the failure thresholds for the frequency response time and the P-value be respectively... and Therefore, when the frequency response time increment compared to the initial value is greater than... Or the increment of the P-value compared to the initial value is greater than If this happens, the system will fail;

[0041] According to equation (7), the lifetime distribution function of the servo controller at normal operating temperature is:

[0042] (8).

[0043] Furthermore, based on the V-optimal criterion, with the optimization objective of minimizing the asymptotic variance of the servo controller's quantile lifetime, a multi-parameter comprehensive optimal criterion is constructed to form the optimization objective function, specifically including:

[0044] Considering that the reliability of the servo controller is determined by both frequency response time and P-value, based on the V-optimal criterion, the goal is to minimize the servo controller's performance. Taking the asymptotic variance of quantile lifetime as the optimization objective, a multi-parameter comprehensive optimal criterion is constructed; the expression of the V-optimal criterion is:

[0045] (9)

[0046] In equation (9), The parameter vector in the multi-parameter accelerated degradation model of the servo controller, as shown in equation (7), , represent The transpose of the matrix, It is the Fisher information matrix of the multi-parameter accelerated degradation model of the servo controller as shown in equation (7). In equation (8) The corresponding probability density function, For servo controllers under normal temperature stress Percentage lifetime;

[0047] for After taking the first-order partial derivatives of equation (8) with respect to each unknown parameter and its combination, we get:

[0048] (10)

[0049] In equation (10), , and They are respectively:

[0050] ;

[0051] ;

[0052] ;

[0053] for According to equation (7) and measurement data The log-likelihood function of the parameters in the multi-parameter accelerated degradation model of the servo controller, as shown in equation (7), is:

[0054] (11)

[0055] Therefore, by taking the second-order partial derivatives of equation (11) with respect to each unknown parameter and its combination, and then taking the negative value, we obtain the Fisher information matrix. The expression is:

[0056] (12)

[0057] In equation (12),

[0058] ;

[0059] ;

[0060] In the above formula, , ;

[0061] Finally, substituting equations (10) and (12) into equation (9) yields the optimized objective function.

[0062] Preferably, determining the experimental design variables specifically includes:

[0063] The goal of the accelerated degradation test design for servo controllers is to determine the test sample size based on a given number of temperature stress levels and test costs. Temperature stress level value and the test sample size at each temperature stress level. Number of measurements and measurement interval Therefore, the design variables of the experiment are denoted as .

[0064] Preferably, the test cost includes at least the cost of the test sample, the measurement cost, and the operating cost of the test equipment; these three types of costs are the main costs involved in accelerated degradation testing. Indicates the cost of a single test sample. Indicates the cost of a single measurement. If the cost of operating the test equipment per unit time is expressed as: Then the total test cost is expressed as:

[0065] (13).

[0066] Preferably, the step of constructing an accelerated degradation test optimization model with the objective function as the objective, and solving the accelerated degradation test optimization model using a genetic algorithm to obtain the servo controller accelerated degradation test scheme specifically includes:

[0067] Given a number of experimental samples Temperature stress level number Under the conditions of experimental cost, combined with the optimization objective function, the accelerated degradation experiment optimization model is obtained as shown in equation (14):

[0068] (14)

[0069] In equation (14), This represents the maximum pre-set test cost;

[0070] Finally, the accelerated degradation test optimization model shown in Equation (14) is solved using a genetic algorithm to obtain the temperature stress level value and the test sample size, number of measurements and measurement interval at each temperature stress level, which is the accelerated degradation test scheme of the servo controller.

[0071] The beneficial effects of this invention are:

[0072] 1. This invention addresses the optimization design of constant temperature stress accelerated degradation tests for servo controllers. A multi-parameter accelerated degradation model for the servo controller is constructed using the Wiener process model, the Arrhenius model, and the Gaussian Copula function. A lifetime distribution function is then derived from this model. Based on the V-optimal criterion and the lifetime distribution function, a multi-parameter comprehensive optimal criterion is constructed to form the optimization objective function. An accelerated degradation test optimization model is then built using this objective function. Finally, a genetic algorithm is used to solve for the optimal temperature stress level and the experimental variables such as sample size, number of measurements, and measurement interval at each temperature stress level. This invention characterizes the coupling relationships between multiple performance parameters, overcoming the limitations of traditional single-performance-parameter design methods and significantly improving the accuracy of predicting servo controller lifetime and reliability based on experimental data.

[0073] 2. The model construction of this invention explicitly considers the experimental cost constraint, and can seek the optimal allocation of experimental resources within a limited budget, thus achieving the best balance between evaluation accuracy and experimental cost.

[0074] 3. The method of the present invention is simple and easy to calculate, and can be directly applied to engineering practice. The evaluation results are stable and reliable, and it is convenient for engineering technicians to learn and master. It is not only applicable to scenarios with two performance parameters, but can also be extended to situations with more performance parameters. It has strong practicality and good application prospects. Attached Figure Description

[0075] Figure 1 This is a schematic flowchart of the accelerated degradation test method for servo controllers based on multi-parameter optimization as described in an embodiment of the present invention. Detailed Implementation

[0076] To facilitate a better understanding of the present invention by those skilled in the art, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The following are merely exemplary and do not limit the scope of protection of the present invention.

[0077] This invention discloses a method for accelerating the degradation test of a servo controller based on multi-parameter optimization, such as... Figure 1 As shown, it includes the following steps:

[0078] Step 1: Construct a multi-parameter accelerated degradation model for the servo controller: Based on the Wiener process model, construct a degradation process model for each performance parameter of the servo controller; use the Arrhenius model to characterize the influence of temperature stress on each performance parameter; use the Gaussian Copula function to characterize the coupling relationship between each performance parameter, thereby obtaining a multi-parameter accelerated degradation model for the servo controller.

[0079] Step 2: Constructing a multi-parameter comprehensive optimal criterion: Based on the multi-parameter accelerated degradation model of the servo controller, derive the lifetime distribution function of the servo controller at normal operating temperature; based on the V-optimal criterion, construct a multi-parameter comprehensive optimal criterion to form the optimization objective function, with minimizing the asymptotic variance of the servo controller's quantile lifetime as the optimization objective.

[0080] Step 3: Construct and solve the accelerated degradation test optimization model: Determine the test design variables, and under the test cost constraint, construct the accelerated degradation test optimization model with the aforementioned optimization objective function as the objective. Use a genetic algorithm to solve the accelerated degradation test optimization model to obtain the servo controller accelerated degradation test scheme.

[0081] The following example uses a certain type of servo controller, combined with... Figure 1 The present invention provides a more detailed description of the accelerated degradation test method for servo controllers based on multi-parameter optimization.

[0082] The operating conditions of a certain type of servo controller are affected by temperature stress, leading to parameter degradation in its 160Hz frequency response time and P-value. Based on engineering experience, the performance degradation process of this type of servo controller is described by the Wiener process, and the effect of temperature on its performance is characterized by the Arrhenius model. Given a test cost of 300,000 yuan and four temperature stress levels, we need to design an accelerated degradation test for multi-parameter optimization of this type of servo controller. The specific steps are as follows:

[0083] Step 1: Construct a multi-parameter accelerated degradation model for the servo controller.

[0084] A joint distribution is constructed using the Wiener process model and the binary Gaussian Copula function. The generalized linear logarithmic model is used as the acceleration model to construct a multi-parameter accelerated degradation model for the servo controller, as shown in Equation (15).

[0085] (15)

[0086] The model parameters estimated using the preliminary test data are as follows: , , , , , , .

[0087] Step 2: Construct a multi-parameter comprehensive optimal criterion.

[0088] Given that the number of stress levels at a constant temperature is 4, denoted as At each group of accelerated temperature stress levels The sample size of the following test is The total experimental sample size is The number of measurements was , The measurement interval is The measured performance degradation parameter is Therefore, the design variables for the experiment are denoted as... .

[0089] According to equation (15), the lifetime distribution function of the servo controller at normal operating temperature is:

[0090] (17)

[0091] In the above formula, let the failure thresholds be respectively %, Normal operating temperature =298.15K.

[0092] The median lifetime can be obtained according to equation (9). The V-optimal criterion takes the following form:

[0093] (16)

[0094] In the above formula, This represents the 0.5 percentile lifetime of the servo controller under normal temperature stress, i.e., the median lifetime. For equation (16)... After taking the first-order partial derivatives of equation (17) with respect to each unknown parameter and its combination, we get:

[0095] (18)

[0096] In equation (18), , and They are respectively:

[0097] ;

[0098] ;

[0099] ;

[0100] For equation (16) The log-likelihood function obtained through equation (15) is:

[0101] (19)

[0102] Therefore, by taking the negative of the expectation of the second-order partial derivatives of the log-likelihood function shown in equation (19) with respect to each unknown parameter and its combination, we obtain the Fisher information matrix. for:

[0103] (20)

[0104] In equation (20),

[0105] ;

[0106] ;

[0107] In the above formula, , .

[0108] Substituting equations (18) and (20) into equation (16), we obtain the optimized objective function.

[0109] Step 3: Construct and solve the accelerated degradation test optimization model.

[0110] After investigation and statistics, the cost of a single test sample is 15,000 yuan, the cost of a single measurement is 200 yuan, the cost of operating the test equipment per unit time is 200 yuan, and the total test cost is 300,000 yuan. According to formula (13), the test cost constraint is:

[0111] (twenty one)

[0112] The normal operating temperature level and extreme temperature level of the servo controller are 298.15K and 363.15K, respectively. Combining the design variables and optimization objective function in step two, the accelerated degradation test optimization model of the servo controller under constant temperature stress can be obtained as shown in equation (22):

[0113] (twenty two)

[0114] The accelerated degradation test optimization model shown in Equation (22) was solved using a genetic algorithm to obtain the optimal temperature stress level value and the test sample size, number of measurements and measurement interval at each temperature stress level, as shown in Table 1 below.

[0115] Table 1 Accelerated Degradation Test Scheme for Servo Controllers

[0116]

[0117] This completes the design of the constant temperature stress accelerated degradation test for this type of servo controller in this embodiment. Test personnel can conduct experiments based on this scheme, and the obtained test data can be used most effectively for high-precision lifespan prediction of the servo controller.

[0118] The above description only outlines the basic principles and preferred embodiments of the present invention. Those skilled in the art can make many changes and modifications based on the above description, and these changes and modifications should fall within the protection scope of the present invention.

Claims

1. A method for accelerating the degradation test of a servo controller based on multi-parameter optimization, characterized in that, Including the following steps: Constructing a multi-parameter accelerated degradation model for a servo controller: Based on the Wiener process model, a degradation process model for each performance parameter of the servo controller is constructed; the influence of temperature stress on each performance parameter is characterized using the Arrhenius model; and the coupling relationship between each performance parameter is characterized using the Gaussian Copula function, thus obtaining a multi-parameter accelerated degradation model for the servo controller. Constructing a multi-parameter comprehensive optimal criterion: Based on the multi-parameter accelerated degradation model of the servo controller, derive the lifetime distribution function of the servo controller at normal operating temperature; based on the V-optimal criterion, construct a multi-parameter comprehensive optimal criterion to form the optimization objective function with minimizing the asymptotic variance of the servo controller's quantile lifetime as the optimization objective; Constructing and solving an accelerated degradation test optimization model: Determine the test design variables, and under the test cost constraint, construct an accelerated degradation test optimization model with the aforementioned optimization objective function as the objective. Use a genetic algorithm to solve the accelerated degradation test optimization model to obtain the servo controller accelerated degradation test scheme.

2. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 1, characterized in that, The performance parameters of a servo controller should include at least frequency response time and P-value.

3. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 1 or 2, characterized in that, The aforementioned model for the degradation process of various performance parameters of the servo controller, constructed based on the Wiener process model, specifically includes: Based on the Wiener process model, a degradation process model of the servo controller is constructed, including two performance parameters: frequency response time and P-value. The expression is as follows: （1） In equation (1), Representing the One performance parameter in The level of degradation at any given moment , Representing the The initial values ​​of each performance parameter, and Representing the first The drift coefficient and diffusion coefficient of the performance parameter degradation process model. Represents standard Brownian motion; No. One performance parameter in Degradation increment at time It follows a normal distribution, that is: The probability density function and cumulative distribution function distribution As shown in equations (2) and (3) respectively: （2） （3） In equation (3), The cumulative distribution function represents the standard normal distribution.

4. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 3, characterized in that, The method of using the Arrennis model to characterize the effect of temperature stress on various performance parameters specifically includes: The Arrhenius model was used to characterize the effects of temperature stress on two performance parameters: frequency response time and P-value. The drift coefficient of the degradation process model was then used as the performance parameter. As a parameter of temperature effect, the constructed acceleration model function is shown in equation (4): （4） In equation (4), Represents absolute temperature. Indicates temperature Next The drift coefficient values ​​of the performance parameter degradation process model. and They respectively represent the corresponding to the first Two parameters of the acceleration model with one performance parameter; The temperature stress is normalized as shown in equation (5): （5） In equation (5), This represents the standardized temperature level value. and These represent the normal operating temperature level and the extreme temperature level of the servo controller, respectively. Therefore, equation (4) can be further expressed as: （6）。 5. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 4, characterized in that, The method of using Gaussian Copula functions to characterize the coupling relationship between various performance parameters to obtain a multi-parameter accelerated degradation model of the servo controller specifically includes: By using the binary Gaussian Copula function to characterize the coupling relationship between the two performance parameters of frequency response time and P-value, the multi-parameter accelerated degradation model of the servo controller is obtained as shown in equation (7): （7） In equation (7), Represents the correlation coefficient. , , This represents the inverse function of the cumulative distribution function of the standard normal distribution. The probability density function represents the joint distribution of the degradation increments of the two performance parameters, frequency response time and P-value.

6. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 5, characterized in that, The derivation of the lifetime distribution function of the servo controller at normal operating temperature based on the multi-parameter accelerated degradation model of the servo controller specifically includes: set up Represents the number of test samples for the servo controller. Represents the temperature stress level number. Representing the A temperature stress level value, Representing the Number of test samples under each temperature stress level Representing the The number of times the performance parameter values ​​are measured under each temperature stress level, and the measurement interval is [number]. ,but Representing the Under the first temperature stress level The sample at the th The first measurement One performance parameter value, Representing the Under the first temperature stress level The drift coefficient of the i-th performance parameter value, the th Temperature stress level After standardization ,in, , , , ; Given that the failure of either the frequency response time or the P-value in the servo controller is considered a system failure, let the failure thresholds for the frequency response time and the P-value be respectively... and Therefore, when the frequency response time increment compared to the initial value is greater than... Or the increment of the P-value compared to the initial value is greater than If this happens, the system will fail; According to equation (7), the lifetime distribution function of the servo controller at normal operating temperature is: （8）。 7. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 6, characterized in that, The aforementioned V-optimal criterion, with minimizing the asymptotic variance of the servo controller's quantile lifetime as the optimization objective, constructs a multi-parameter comprehensive optimal criterion to form the optimization objective function, specifically including: Based on the V-optimal criterion, to minimize the servo controller Taking the asymptotic variance of quantile lifetime as the optimization objective, a multi-parameter comprehensive optimal criterion is constructed; the expression of the V-optimal criterion is: （9） In equation (9), The parameter vector in the multi-parameter accelerated degradation model of the servo controller, as shown in equation (7), , It is the Fisher information matrix of the multi-parameter accelerated degradation model of the servo controller as shown in equation (7). In equation (8) The corresponding probability density function, For servo controllers under normal temperature stress Percentage lifetime; for After taking the first-order partial derivatives of equation (8) with respect to each unknown parameter and its combination, we get: （10） In equation (10), , and They are respectively: ； ； ； for According to equation (7) and measurement data The log-likelihood function of the parameters in the multi-parameter accelerated degradation model of the servo controller, as shown in equation (7), is: （11） Therefore, by taking the second-order partial derivatives of equation (11) with respect to each unknown parameter and its combination, and then taking the negative value, we obtain the Fisher information matrix. The expression is: （12） In equation (12), ； ； In the above formula, , ; Finally, substituting equations (10) and (12) into equation (9) yields the optimized objective function.

8. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 7, characterized in that, The determination of experimental design variables specifically includes: Determine the experimental sample size Temperature stress level value and the test sample size at each temperature stress level. Number of measurements and measurement interval Therefore, the design variables of the experiment are denoted as .

9. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 8, characterized in that, The test costs include at least the cost of the test samples, measurement costs, and operating costs of the test equipment; Indicates the cost of a single test sample. Indicates the cost of a single measurement. If the cost of operating the test equipment per unit time is expressed as: Then the total test cost is expressed as: （13）。 10. The accelerated degradation test method for servo controllers based on multi-parameter optimization according to claim 9, characterized in that, The process involves constructing an accelerated degradation test optimization model with the objective function as the target, and solving the accelerated degradation test optimization model using a genetic algorithm to obtain a servo controller accelerated degradation test scheme, specifically including: Given a number of experimental samples Temperature stress level number Under the conditions of experimental cost, combined with the optimization objective function, the accelerated degradation experiment optimization model is obtained as shown in equation (14): （14） In equation (14), This represents the preset maximum test cost; Finally, the accelerated degradation test optimization model shown in Equation (14) is solved using a genetic algorithm to obtain the temperature stress level value and the test sample size, number of measurements and measurement interval at each temperature stress level, which is the accelerated degradation test scheme of the servo controller.

Images