Capacitor service life detection method, device, equipment and medium

Abstract

The invention relates to the technical field of capacitor life detection. The capacitor service life detection method comprises the steps of obtaining electrical performance response information of a capacitor under a multi-excitation condition, constructing a dynamic response behavior path of the capacitor in a state space based on the electrical performance response information, obtaining an evolution behavior sequence, and calculating the service life of the capacitor according to the evolution behavior sequence. And matching the evolution behavior sequence with the aging behavior path in the aging trajectory library, identifying a corresponding degradation trend type, obtaining current detection condition information, inputting the degradation trend type into a life evaluation model, and performing analysis in combination with the current detection condition information to obtain a life evaluation result. The method has the effect of improving the accuracy of service life evaluation of the capacitor.

Description

Technical Field

[0001] The present invention relates to the technical field of capacitor life detection, and in particular to a capacitor life detection method, device, equipment and medium. Background Art

[0002] As key components in electronic equipment, capacitors are widely used in power systems, communications equipment, and consumer electronics. The stability of their performance directly impacts the overall system's operating efficiency and reliability. As modern industry continues to demand more stability and safety from equipment, capacitor performance monitoring and lifespan assessment have become crucial for ensuring long-term, stable operation. This not only helps reduce unexpected failures but also effectively reduces maintenance costs and improves overall economic benefits.

[0003] In existing technology, capacitor lifespan assessment typically involves the following methods: First, determining the degree of aging by regularly monitoring changes in the capacitor's capacitance; second, indirectly inferring the capacitor's condition by analyzing the changing patterns of environmental parameters such as temperature or pressure; and third, assessing its health by analyzing the byproducts of chemical reactions within the capacitor. While these methods can reflect the aging trend of capacitors to a certain extent, they all focus on a single factor and lack a holistic analysis of the capacitor's multi-dimensional dynamic response behavior.

[0004] However, the above conventional methods generally have the following defects: they are unable to fully capture the dynamic response characteristics of capacitors under complex excitation conditions, and it is difficult to accurately identify their degradation trend types, resulting in insufficient accuracy of life assessment results. Summary of the Invention

[0005] In order to improve the accuracy of capacitor life assessment, the present application provides a capacitor life detection method, device, equipment and medium.

[0006] The above-mentioned invention objective of this application is achieved through the following technical solutions: A capacitor life detection method, comprising: Obtaining the electrical performance response information of the capacitor under multiple excitation conditions; Based on the electrical performance response information, constructing a dynamic response behavior path of the capacitor in a state space to obtain an evolutionary behavior sequence; Matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type; Get current detection condition information; The degradation trend type is input into the life assessment model and analyzed in combination with the current detection condition information to obtain a life assessment result.

[0007] By adopting the above technical solution, by acquiring the electrical performance response information of the capacitor under multiple excitation conditions, it is possible to comprehensively capture the operating characteristics of the capacitor under different conditions such as frequency, voltage, and temperature, avoiding one-sided information under a single operating condition, thereby improving the perception of the capacitor's operating status and potential aging behavior. By constructing a dynamic response behavior path in the state space based on the electrical performance response information, multiple electrical performance parameters can be jointly mapped into a time series trajectory, reflecting the overall evolution trend of the capacitor's performance as the excitation conditions change, thereby enhancing the modeling accuracy of the capacitor's aging process. By matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library, the response characteristics obtained by real-time detection can be associated with the existing aging model, thereby quickly determining the degradation trend of the capacitor without relying on long-term operation tests. By obtaining current detection condition information and inputting it into the life assessment model for analysis, the aging process can be modified based on actual operating conditions, thereby outputting a remaining life prediction result that better fits the field application conditions, improving the accuracy and practicality of life assessment.

[0008] In a preferred example, the present application may be further configured as follows: constructing a dynamic response behavior path of the capacitor in a state space based on the electrical performance response information to obtain an evolutionary behavior sequence, including: Preprocessing the electrical performance response information to obtain preprocessed multi-dimensional response parameters; Mapping the preprocessed multidimensional response parameters into the state space to form a state point sequence; The state point sequence is arranged in the order of changes in the incentive conditions, and a state path structure is established with the incentive dimension as an index to generate the evolutionary behavior sequence.

[0009] By adopting the above technical solution and preprocessing the electrical performance response information, it is possible to eliminate outliers in the acquisition process, fill in missing items, smooth short-term fluctuations, and normalize parameter values of different dimensions, thereby improving the quality and comparability of the original data and ensuring the stability of the subsequent calculation process. By mapping the preprocessed multidimensional response parameters to the state space, it is possible to express multidimensional parameters such as capacitance value, equivalent series resistance, and loss tangent value in a geometric manner as state points, and construct a set of state points suitable for spatial modeling, thereby providing a structural basis for subsequent trajectory construction. By arranging the state points in the order of changes in the excitation conditions and establishing a state path structure with the excitation dimension as the index, it is possible to truly restore the continuous trajectory of the capacitor response behavior as the excitation changes.

[0010] In a preferred example, the present application can be further configured as follows: arranging the state point sequence in the order of change of the incentive conditions, and establishing a state path structure with the incentive dimension as an index to generate the evolutionary behavior sequence, including: Using the frequency value in the excitation dimension as the index axis, sorting the state point sequence in ascending order of frequency to obtain a sorted state point sequence; Based on the sorted state point sequence, a connection relationship between the state points is established to construct the evolution behavior sequence.

[0011] By adopting the above technical solution, by taking the frequency value in the excitation dimension as the index axis and sorting the state point sequence in ascending order of frequency, the evolution direction of the behavior trajectory can be kept consistent with the change of the excitation parameters, thereby enhancing the temporal logical consistency and interpretability of the trajectory; by establishing the connection relationship between the state points based on the sorted state point sequence, a path structure with continuity and directionality can be constructed, thereby effectively describing the evolution path of the capacitor response characteristics in the state space and forming an evolutionary behavior sequence for subsequent trend comparison.

[0012] In a preferred example, the present application may be further configured as follows: matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type includes: Calculating the similarity between the evolutionary behavior sequence and the aging behavior path in the aging trajectory library to obtain a similarity score; Based on the similarity score, a matching aging behavior path is confirmed, and according to the matching aging behavior path, the corresponding degradation trend type is identified.

[0013] By adopting the above technical solution, by calculating the similarity between the evolutionary behavior sequence and the aging behavior path in the aging trajectory library, it is possible to quantitatively evaluate the degree of proximity between the current response mode of the capacitor and the known aging pattern, thereby realizing intelligent identification of the current aging state of the capacitor; by confirming the most matching aging behavior path based on the similarity score and extracting the corresponding degradation trend type, it is possible to quickly match the object to be tested with the standard trend type, thereby replacing manual experience judgment or long-term monitoring process, and improving the degree of automation and efficiency of trend identification.

[0014] In a preferred example, the present application may be further configured as follows: calculating the similarity between the evolutionary behavior sequence and the aging behavior path in the aging trajectory library to obtain a similarity score, including: Performing dimension normalization processing on the corresponding state nodes in the evolutionary behavior sequence and the aging behavior path to obtain a normalized state point sequence pair; Based on the normalized state point sequence pairs, calculating the Euclidean distance between corresponding state nodes to obtain a distance set; The distance set is weightedly accumulated to obtain the similarity score.

[0015] By adopting the above technical solution, by dimensional normalizing the corresponding state nodes in the evolutionary behavior sequence and the aging behavior path, the influence of the numerical scale differences between different parameters on the distance calculation can be eliminated, thereby ensuring that the similarity calculation is carried out under a unified standard and improving the objectivity of the calculation results; by calculating the Euclidean distance between each corresponding state node based on the normalized state point sequence, the degree of local deviation between each group of state points can be accurately characterized, thereby establishing a fine-grained quantitative basis for behavioral differences; by weighted accumulation of the distance set, the contribution of different state point pairs in the matching can be considered, and an overall similarity score can be output, thereby providing a quantitative basis with controllable accuracy for trend matching and degradation identification.

[0016] In a preferred example, the present application may be further configured as follows: the degradation trend type is input into the life assessment model, and analyzed in combination with the current detection condition information to obtain a life assessment result, including: By selecting a corresponding life evolution function for the degradation trend type in the life assessment model; Substituting the current detection condition information as an input variable into the life evolution function to obtain a life response curve; Performing trend analysis on the life response curve to obtain the life evaluation result.

[0017] By adopting the above technical solution, by inputting the degradation trend type into the life assessment model and selecting the corresponding life evolution function, it is possible to ensure that the prediction model used is highly consistent with the actual aging path of the capacitor, thereby improving the accuracy of life assessment; by substituting the current detection condition information as the input variable into the life evolution function, the influence of factors such as ambient temperature and voltage stress on the aging rate can be considered, thereby realizing dynamic life reasoning for working condition adaptation; by performing trend analysis on the life response curve, the inflection point, slope and downward trend in the life change process can be determined, thereby more accurately estimating the remaining useful life of the capacitor and improving the interpretability and engineering applicability of the prediction results.

[0018] In a preferred example, the present application may be further configured as follows: performing trend analysis on the life response curve to obtain the life assessment result includes: Fitting the life response curve based on the least square method to generate a life change fitting function; Solving the intersection of the life change fitting function using a preset life determination threshold to obtain the horizontal coordinate of the intersection of the preset life determination threshold and the life change fitting function; The time value corresponding to the abscissa of the intersection is used as the life evaluation result of the capacitor.

[0019] By adopting the above technical solution, by fitting the life response curve based on the least squares method, the discrete response data can be fitted into a continuous, smooth and trend-stable mathematical expression, which facilitates subsequent analytical processing and curve trend research; by setting the life judgment threshold and solving the intersection of the fitting function, the time position corresponding to the capacitor life falling to the failure standard can be determined, thereby achieving accurate and calculable life end prediction; by using the horizontal coordinate of the intersection as the life assessment result, the life value corresponding to the actual operating time can be directly output, thereby providing a quantitative basis for capacitor maintenance decisions.

[0020] The second object of the present invention is achieved through the following technical solutions: A capacitor life detection device, comprising: Electrical performance data acquisition module, used to obtain the electrical performance response information of the capacitor under multiple excitation conditions; A behavior trajectory modeling module is used to construct a dynamic response behavior path of the capacitor in a state space based on the electrical performance response information to obtain an evolutionary behavior sequence; a trend identification module, configured to match the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type; Environmental parameter acquisition module, used to obtain current detection condition information; The life assessment analysis module is used to input the degradation trend type into the life assessment model, and analyze it in combination with the current detection condition information to obtain a life assessment result.

[0021] By adopting the above technical solution, by acquiring the electrical performance response information of the capacitor under multiple excitation conditions, it is possible to comprehensively capture the operating characteristics of the capacitor under different conditions such as frequency, voltage, and temperature, avoiding one-sided information under a single operating condition, thereby improving the perception of the capacitor's operating status and potential aging behavior. By constructing a dynamic response behavior path in the state space based on the electrical performance response information, multiple electrical performance parameters can be jointly mapped into a time series trajectory, reflecting the overall evolution trend of the capacitor's performance as the excitation conditions change, thereby enhancing the modeling accuracy of the capacitor's aging process. By matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library, the response characteristics obtained by real-time detection can be associated with the existing aging model, thereby quickly determining the degradation trend of the capacitor without relying on long-term operation tests. By obtaining current detection condition information and inputting it into the life assessment model for analysis, the aging process can be modified based on actual operating conditions, thereby outputting a remaining life prediction result that better fits the field application conditions, improving the accuracy and practicality of life assessment.

[0022] The third objective of this application is achieved through the following technical solutions: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned capacitor life detection method are implemented.

[0023] The fourth objective of this application is achieved through the following technical solutions: A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned capacitor life detection method.

[0024] In summary, this application includes at least one of the following beneficial technical effects: 1. By acquiring the electrical performance response information of capacitors under multiple excitation conditions, it is possible to comprehensively capture the operating characteristics of capacitors under different frequency, voltage, and temperature conditions, avoiding one-sided information under a single operating condition, thereby improving the ability to perceive the working status and potential aging behavior of the capacitor. By constructing a dynamic response behavior path in the state space based on the electrical performance response information, multiple electrical performance parameters can be jointly mapped into a time series trajectory, reflecting the overall evolution trend of capacitor performance as the excitation conditions change, thereby enhancing the modeling accuracy of the capacitor aging process. By matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library, the response characteristics obtained by real-time detection can be associated with the existing aging model, thereby quickly determining the degradation trend of the capacitor without relying on long-term operation tests. By obtaining the current detection condition information and inputting it into the life assessment model for analysis, the aging process can be modified in combination with the actual operating conditions, thereby outputting a remaining life prediction result that better fits the field application conditions, improving the accuracy and practicality of life assessment. 2. By inputting the degradation trend type into the life assessment model and selecting the corresponding life evolution function, it is possible to ensure that the prediction model used is highly consistent with the actual aging path of the capacitor, thereby improving the accuracy of life assessment. By substituting the current detection condition information as the input variable into the life evolution function, the impact of factors such as ambient temperature and voltage stress on the aging rate can be considered, thereby achieving dynamic life reasoning for working condition adaptation. By performing trend analysis on the life response curve, the inflection point, slope and downward trend in the life change process can be determined, thereby more accurately estimating the remaining useful life of the capacitor and improving the interpretability and engineering applicability of the prediction results. 3. By fitting the life response curve based on the least squares method, the discrete response data can be fitted into a continuous, smooth and trend-stable mathematical expression, which facilitates subsequent analytical processing and curve trend research; by setting the life judgment threshold and solving the intersection of the fitting function, the time position corresponding to the capacitor life falling to the failure standard can be determined, thereby achieving accurate and calculable life end prediction; by using the horizontal coordinate of the intersection as the life assessment result, the life value corresponding to the actual operating time can be directly output, thereby providing a quantitative basis for capacitor maintenance decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic structural diagram of a capacitor life detection method according to an embodiment of the application; Figure 2 This is a flowchart for implementing step S20 in a capacitor life detection method in one embodiment of the present application; Figure 3 This is a flowchart for implementing step S203 in a capacitor life detection method in one embodiment of the present application; Figure 4 This is a flowchart for implementing step S30 in a capacitor life detection method in one embodiment of the present application; Figure 5 This is a flowchart for implementing step S301 in a capacitor life detection method in one embodiment of the present application; Figure 6 This is a flowchart for implementing step S50 in a capacitor life detection method in one embodiment of the present application; Figure 7 This is a flowchart for implementing step S503 in a capacitor life detection method in one embodiment of the present application; Figure 8 This is a principle block diagram of a capacitor life detection device in one embodiment of the present application; Figure 9 It is a schematic diagram of a device in one embodiment of the present application. DETAILED DESCRIPTION

[0026] The present application is further described in detail below with reference to the accompanying drawings.

[0027] In one embodiment, if Figure 1 As shown, the present application discloses a capacitor life detection method, which specifically includes the following steps: S10: Obtaining electrical performance response information of the capacitor under multiple excitation conditions.

[0028] In this embodiment, multiple excitation conditions refer to applying a combination of excitations including different frequencies, voltage amplitudes, and ambient temperatures to the capacitor. By adjusting the excitation source parameters and the temperature control device, the capacitor produces an electrical performance response under multiple typical operating conditions. The response information includes parameters such as the capacitance value, equivalent series resistance (ESR), and loss tangent value measured under each excitation condition.

[0029] Specifically, when obtaining the electrical performance response information of the capacitor under multiple excitation conditions, first establish a connection between the capacitor to be tested and the test device, and set a combination of excitation conditions of different frequencies, voltage amplitudes and ambient temperatures. The frequency can be set to multiple frequency points such as 10Hz, 100Hz, 1kHz, etc. The voltage amplitude can be set to different levels of 70%, 90% and 100% according to the rated voltage of the capacitor. The ambient temperature is adjusted to 25°C, 60°C or 85°C through the temperature control cavity. Each set of excitation conditions is applied to both ends of the capacitor respectively, and the response data of the capacitor under each set of excitation conditions is obtained through an impedance analyzer or an equivalent measuring device. The capacitance value can be calculated by integrating the voltage-time curve during the charging and discharging process, the equivalent series resistance is obtained by measuring the real part of the impedance under high-frequency conditions and correcting the wire resistance, and the loss tangent value is obtained by measuring the angle between the capacitive reactance component of the capacitor and the total impedance. The collected data is organized in the excitation sequence to form complete electrical performance response information under multiple excitation points. S20: Based on the electrical performance response information, construct a dynamic response behavior path of the capacitor in a state space to obtain an evolutionary behavior sequence.

[0030] In this embodiment, the state space refers to a parameter space constructed with the electrical performance response parameters of the capacitor as multi-dimensional coordinate axes, and the electrical performance response parameters include capacitance value, equivalent series resistance, loss tangent value, etc.

[0031] Specifically, when constructing the dynamic response behavior path of the capacitor in the state space based on the electrical performance response information, the collected capacitance value, equivalent series resistance and loss tangent value are first regularized, and the response parameters under different excitation conditions are uniformly mapped to the multi-dimensional state space, where the capacitance value is used as the first-dimensional coordinate axis, the equivalent series resistance is used as the second-dimensional coordinate axis, and the loss tangent value is used as the third-dimensional coordinate axis. Each set of response parameters corresponds to a state point in the state space. Then, the state points are arranged in the order of setting the excitation conditions, and the adjacent state points are connected in pairs according to the arrangement order to form a trajectory path. The entire trajectory path constitutes the response behavior trajectory of the capacitor in the state space. The trajectory shows a continuous change trend under different excitation conditions, and fully records the parameter response distribution process of the capacitor under the current stage. The constructed trajectory path is then output and recorded in the form of a data sequence to form an evolutionary behavior sequence.

[0032] S30: Matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type.

[0033] In this embodiment, the aging trajectory library refers to a pre-established trajectory set for storing the dynamic response behavior paths of typical capacitor aging processes. The trajectories are constructed based on historical experimental data, simulation models, or field operation monitoring data, and represent multiple known degradation trend types.

[0034] Specifically, when matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library, multiple classified aging behavior path samples are first extracted from the aging trajectory library. The sample paths are established through historical experimental records, long-term operation monitoring data or simulation modeling. Each path consists of a set of state point sequences, which are used to characterize the response trajectory pattern of a specific type of capacitor during the aging process. The state point sequence in the evolutionary behavior sequence is aligned with each aging behavior path, and the aligned point pairs are normalized to eliminate the differences in the numerical magnitude of response parameters in different dimensions. Then, the least squares fitting or distance measurement method is used to calculate the similarity between the evolutionary behavior sequence and each aging behavior path. The similarity calculation includes summing the Euclidean distance between each pair of normalized state points, and using the cumulative distance value as the fitting error index between the trajectories. Finally, the degradation type corresponding to the aging behavior path with the smallest fitting error is selected as the degradation trend type of the current capacitor.

[0035] S40: Obtain current detection condition information.

[0036] Specifically, the temperature value of the current environment is collected by reading the output signal of the temperature sensor, and the temperature signal is converted into standard format data that can be used for analysis using an analog signal conversion circuit. The real-time amplitude of the applied voltage is detected through a voltage acquisition channel connected to the input end of the capacitor, and the average level of the voltage value is calculated within the set sampling period. The current operating frequency is determined by recording the output frequency of the excitation signal source, and the acquired temperature, voltage and frequency data are synchronously integrated to form a parameter group with a consistent structure to describe the working status during detection, so as to obtain the current detection condition information.

[0037] S50: Inputting the degradation trend type into a lifespan assessment model and analyzing it in combination with the current detection condition information to obtain a lifespan assessment result.

[0038] In this embodiment, the life assessment model refers to a modeling and analysis structure for estimating the remaining life of a capacitor based on the degradation trend type of the capacitor and current detection condition information.

[0039] Specifically, when the degradation trend type is input into the life assessment model and analyzed in combination with the current detection condition information, the corresponding life calculation function is selected according to the mapping relationship between the preset trend type and the life model. The life calculation function is used to describe the numerical relationship between the life index and the environmental parameters of the capacitor under a specific aging trend. The temperature, voltage and frequency values in the current detection conditions are substituted into the life calculation expression in sequence as input variables of the function. The calculation process is executed by the substituted function to obtain the life response result curve, and then the horizontal axis time value is extracted according to the intersection position between the life curve and the set judgment threshold. The remaining life estimate of the capacitor under the current working conditions is inferred through this time value to obtain the life assessment result.

[0040] In one embodiment, if Figure 2 As shown, in step S20, that is, based on the electrical performance response information, a dynamic response behavior path of the capacitor in the state space is constructed to obtain an evolutionary behavior sequence, including: S201: Preprocessing the electrical performance response information to obtain preprocessed multi-dimensional response parameters.

[0041] Specifically, when preprocessing the electrical performance response information, the collected capacitance, equivalent series resistance and loss tangent data are first checked for integrity, and missing items and outliers caused by communication interruption or sensing error are eliminated. Then, different parameters are normalized so that the data of each dimension are in the same numerical scale range. During the normalization process, the minimum-maximum transformation method is used to map the original value to between 0 and 1. Then, the short-term jitter noise in the sequence is filtered and smoothed using the sliding average algorithm to improve the continuity and analysis stability of the response data. Finally, a set of preprocessed capacitance values, equivalent series resistance and loss tangent values corresponding to each excitation condition are output to obtain the preprocessed multidimensional response parameters.

[0042] S202: Mapping the pre-processed multi-dimensional response parameters into the state space to form a state point sequence.

[0043] Specifically, when mapping the preprocessed multidimensional response parameters to the state space, the capacitance value in each set of parameters is set as the first-dimensional coordinate of the state space, the equivalent series resistance is used as the second-dimensional coordinate, and the loss tangent value is used as the third-dimensional coordinate. Each set of parameters corresponds to a state point in the state space. During the mapping process, the state point coordinates are generated in sequence according to the parameter positions. The correspondence of the excitation conditions is recorded by coordinate annotation. Each state point has a unique position in the state space. All state points are arranged in sequence and combined to form a preliminary state point sequence to form a state point sequence.

[0044] S203: Arrange the state point sequence according to the change order of the incentive conditions, and establish a state path structure with the incentive dimension as an index to generate the evolution behavior sequence.

[0045] Specifically, when arranging the state point sequence according to the change order of the excitation conditions, all state points are sorted in the ascending order of the frequency in the set excitation parameters. During the sorting process, the binding relationship between the excitation conditions and each state point is maintained. After the sorting is completed, the adjacent state points are connected in sequence starting from the first state point to form path segments. The connection order is consistent with the excitation conditions. All path segments are connected in series to construct a continuous state path structure. In the state path structure, the frequency is used as the index axis to identify the state point position and change direction. The structure is used to express the change trajectory of the capacitor response in the state space to generate an evolutionary behavior sequence.

[0046] In one embodiment, if Figure 3 As shown, in step S203, the state point sequence is arranged in the order of change of the incentive conditions, and a state path structure is established with the incentive dimension as an index to generate the evolution behavior sequence, including: S2031: Using the frequency value in the excitation dimension as the index axis, sorting the state point sequence in ascending order of frequency to obtain a sorted state point sequence.

[0047] Specifically, when the frequency value in the excitation dimension is used as the index axis to sort the state point sequence, the frequency information corresponding to each state point is first extracted, and an index set containing the correspondence between the state point and the frequency value is constructed. Then, the state points are rearranged in order from low to high according to the frequency value, and the sorting result is used as the new state point arrangement order. At the same time, the mapping relationship between each state point and its original excitation condition is retained to avoid information dislocation during the sorting process. After the sorting is completed, a state point sequence arranged in a monotonically increasing order according to the frequency is generated to obtain the sorted state point sequence.

[0048] S2032: Based on the sorted state point sequence, establish a connection relationship between state points to construct the evolutionary behavior sequence.

[0049] Specifically, when establishing a connection relationship based on the sorted state point sequence, starting from the first state point in the sorted sequence, the current point and its next adjacent point are selected in sequence, and a directed connecting line segment is constructed between the two state points using the continuity of the coordinate position to represent the changing path of the capacitor response state at adjacent frequencies. This operation is repeated until the last state point to form a complete path connection structure, and all connecting line segments are combined to represent a sequential response trajectory curve, which is the dynamic evolution path of the capacitor in the state space, so as to construct the evolutionary behavior sequence.

[0050] In one embodiment, if Figure 4 As shown, in step S30, matching the evolution behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type includes: S301: Calculating the similarity between the evolutionary behavior sequence and the aging behavior path in the aging trajectory library to obtain a similarity score.

[0051] Specifically, when calculating the similarity between the evolutionary behavior sequence and the aging behavior path in the aging trajectory library, the state point sequence in the evolutionary behavior sequence and the state point sequence in each aging behavior path are first dimensionally aligned to ensure that each pair of state points are comparable in parameter dimension and order. After the alignment is completed, the multidimensional parameters in each pair of corresponding state points are normalized separately, and the original values are mapped to a unified numerical range. After normalization, the Euclidean distance of each pair of state points in the state space is calculated as the local response difference. The distance values of all state point pairs are accumulated in turn to form an overall response difference measurement result. Finally, the accumulated total distance value is output as the similarity score, which is used to measure the degree of match between the evolutionary behavior sequence and each aging behavior path to obtain a similarity score.

[0052] S302: confirming a matching aging behavior path based on the similarity score, and identifying the corresponding degradation trend type according to the matching aging behavior path.

[0053] Specifically, when confirming the matching aging behavior path based on the similarity score, all scoring results are sorted from low to high according to the numerical value, and the aging behavior path with the lowest similarity score is selected as the most matching path. The degradation trend type label corresponding to the path is read. This label is a standard identifier attached according to the trajectory source, aging mechanism and actual manifestation during the trajectory library construction phase. This identifier is used as the basis for judging the degradation trend of the current evolutionary behavior sequence to confirm the degradation trend type identification result of the capacitor.

[0054] In one embodiment, if Figure 5 As shown, in step S301, the similarity between the evolution behavior sequence and the aging behavior path in the aging trajectory library is calculated to obtain a similarity score, including: S3011: Performing dimension normalization processing on the evolution behavior sequence and the corresponding state nodes in the aging behavior path to obtain a normalized state point sequence pair.

[0055] Specifically, when performing dimensional normalization processing on the corresponding state nodes in the evolutionary behavior sequence and the aging behavior path, first, multiple response parameter values contained in each pair of corresponding state nodes are extracted, and the original parameter matrix is constructed. The capacitance value, equivalent series resistance and loss tangent value are extracted as independent parameter columns, and each column parameter is linearly transformed using the minimum-maximum normalization method. The minimum value of the column is subtracted from each original parameter value and then divided by the difference between the maximum and minimum values of the column. After completing the normalized mapping, the new state point vectors are recombined to finally obtain a normalized state point sequence pair within a unified scale range.

[0056] S3012: Based on the normalized state point sequence pairs, calculate the Euclidean distance between corresponding state nodes to obtain a distance set.

[0057] Specifically, when calculating the Euclidean distance between each corresponding state node in the normalized state point sequence pair, the three-dimensional coordinate components of each pair of normalized state points are extracted respectively, the difference of each dimensional coordinate is squared and summed up, and then the square root of the sum is taken to obtain the Euclidean distance value between the two state points. This operation is repeated to traverse all corresponding state nodes in the entire state point sequence pair, and the Euclidean distance results calculated for each point pair are recorded in turn. Finally, all Euclidean distance values are organized into an ordered set to obtain a distance set.

[0058] S3013: Perform weighted accumulation on the distance set to obtain the similarity score.

[0059] Specifically, when weighted accumulation is performed on the distance set, the distance weight value between each state point is first set. The weight can be assigned based on different stages of the excitation frequency, the sensitivity of parameter changes or empirical rules. Each Euclidean distance value is multiplied by the corresponding weight coefficient to obtain the weighted distance result. All weighted distance values are then accumulated in sequence order to obtain the overall matching error value. The smaller the error value, the higher the similarity between the behavior sequences. Finally, the weighted error value is used as the similarity score to obtain the similarity score.

[0060] In one embodiment, if Figure 6 As shown, in step S50, the degradation trend type is input into the life assessment model, and analyzed in combination with the current detection condition information to obtain a life assessment result, including: S501: Selecting a corresponding life evolution function according to the degradation trend type in the life assessment model.

[0061] Specifically, when the degradation trend type is used to select the life evolution function in the life assessment model, the life function type identifier corresponding to the current degradation trend type is retrieved according to the pre-established mapping relationship between the trend type and the life function, and the mathematical expression matching the identifier is retrieved from the model function library. The life evolution function is used to characterize the life change process of the capacitor under this type of degradation trend. The function form may include a linear function, an exponential decay function or a piecewise combination function. Different function forms correspond to different aging characteristics. During the function selection process, its number, parameter template and function structure are recorded to obtain the life evolution function to be called.

[0062] S502: Substituting the current detection condition information as an input variable into the life evolution function to obtain a life response curve.

[0063] Specifically, when the current detection condition information is substituted into the life evolution function as an input variable, the environmental parameter values such as temperature, voltage and frequency are substituted into the corresponding variable positions in the life function expression respectively, and the numerical values are input in order according to the variable symbols and parameter structure preset in the function. The variable replacement and function evaluation operations are completed in sequence through the calculation engine, and discrete sampling is performed within the entire parameter range to form a continuous function image point series, and a function curve of the life response changing with the input conditions is drawn. This curve is used to express the life performance trend of the capacitor under the current working conditions to obtain the life response curve.

[0064] S503: Perform trend analysis on the life response curve to obtain the life assessment result.

[0065] Specifically, when performing trend analysis on the life response curve, first scan all vertical axis life values on the curve image to confirm whether there are characteristic forms such as monotonically decreasing, critical turning points or accelerated attenuation. Then, based on the failure judgment threshold set in the model, calculate the intersection position of the curve and the threshold horizontal line, and extract the horizontal axis variable value corresponding to the intersection. The variable value represents the operating time or number of load cycles required to reach the end of life. The remaining life estimation result is inferred from the horizontal axis value as the life assessment indicator of the capacitor under the current detection state to obtain the life assessment result.

[0066] In one embodiment, if Figure 7 As shown, in step S503, that is, performing trend analysis on the life response curve to obtain the life assessment result, including: S5031: Fitting the life response curve based on the least squares method to generate a life change fitting function.

[0067] In this embodiment, the least squares method refers to a mathematical calculation method for fitting a set of discrete data points. By constructing an objective function containing a residual term, the sum of squares of the residuals is minimized to obtain a function expression that best conforms to the changing trend of the original data.

[0068] Specifically, when fitting the life response curve based on the least squares method, first select multiple discrete sampling points on the curve image to form the original data set, use the horizontal axis as the independent variable vector, and the vertical axis life value as the dependent variable vector, establish a set of weighted residual expressions, use the least squares calculation method to minimize the sum of squared residuals, perform coefficient solving operations to obtain the optimal fitting curve function expression in polynomial or exponential form, and after fitting is completed, substitute the original data points into the function to test the fitting accuracy, confirm that the function curve covers the trend consistency and has continuity and smoothness, so as to generate a life change fitting function.

[0069] S5032: Solving the intersection of the life variation fitting function using a preset life determination threshold value to obtain the horizontal coordinate of the intersection of the preset life determination threshold value and the life variation fitting function.

[0070] In this embodiment, the preset life determination threshold refers to a fixed life indicator value used to determine whether the capacitor has reached a failure state.

[0071] Specifically, when solving the intersection of the life change fitting function, first set a life judgment threshold as a fixed reference line. The threshold corresponds to the judgment benchmark of life failure. Substitute the threshold as a constant term into the life change fitting function expression, construct an equation group so that the fitting function is equal to the constant term, and obtain the horizontal coordinate value of the intersection position of the fitting function and the judgment threshold by solving the equation group. The horizontal coordinate represents the time variable value corresponding to when the life drops to the threshold level, so as to obtain the horizontal coordinate of the intersection point of the life change fitting function and the life judgment threshold.

[0072] S5033: Use the time value corresponding to the horizontal coordinate of the intersection as the life evaluation result of the capacitor.

[0073] Specifically, when the horizontal coordinate of the intersection is used as the capacitor life assessment result, it is confirmed that the horizontal coordinate corresponds to the independent variable time axis in the life response function, which represents the remaining operating time or number of service cycles of the capacitor under the current working conditions from the current detection moment. The horizontal coordinate value is converted into an actual time unit expression, such as hours or days, as the final output value for judging the life status to obtain the capacitor life assessment result.

[0074] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0075] In one embodiment, a capacitor life detection device is provided, which corresponds one-to-one to a capacitor life detection method in the above embodiment. Figure 8 As shown, the capacitor life detection device includes an electrical performance data acquisition module, a behavior trajectory modeling module, a trend identification module, an environmental parameter acquisition module, and a life assessment and analysis module. The functional modules are described in detail as follows: Electrical performance data acquisition module, used to obtain the electrical performance response information of the capacitor under multiple excitation conditions; A behavior trajectory modeling module is used to construct a dynamic response behavior path of the capacitor in a state space based on the electrical performance response information to obtain an evolutionary behavior sequence; a trend identification module, configured to match the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type; Environmental parameter acquisition module, used to obtain current detection condition information; The life assessment analysis module is used to input the degradation trend type into the life assessment model, and analyze it in combination with the current detection condition information to obtain a life assessment result.

[0076] Optionally, the behavior trajectory modeling module includes: A response parameter preprocessing submodule, configured to preprocess the electrical performance response information to obtain preprocessed multi-dimensional response parameters; A state point construction submodule, configured to map the preprocessed multidimensional response parameters into the state space to form a state point sequence; The behavior path generation submodule is used to arrange the state point sequence according to the change order of the incentive conditions, establish a state path structure with the incentive dimension as the index, and generate the evolution behavior sequence.

[0077] Optionally, the behavior path generation submodule includes: an excitation index sorting unit, configured to use the frequency value in the excitation dimension as the index axis and sort the state point sequence in ascending order of frequency to obtain a sorted state point sequence; The state path construction unit is used to establish a connection relationship between state points based on the sorted state point sequence to construct the evolution behavior sequence.

[0078] Optional trend identification modules include: a behavior trajectory comparison submodule, configured to calculate the similarity between the evolutionary behavior sequence and the aging behavior path in the aging trajectory library to obtain a similarity score; The trend type identification submodule is configured to confirm a matching aging behavior path based on the similarity score, and identify the corresponding degradation trend type according to the matching aging behavior path.

[0079] Optionally, the behavior trajectory comparison submodule includes: a parameter normalization processing unit, configured to perform dimension normalization processing on the corresponding state nodes in the evolutionary behavior sequence and the aging behavior path to obtain a normalized state point sequence pair; A state distance calculation unit, configured to calculate the Euclidean distance between corresponding state nodes based on the normalized state point sequence pairs to obtain a distance set; The similarity score generating unit is used to perform weighted accumulation on the distance set to obtain the similarity score.

[0080] Optional lifespan assessment analysis modules include: A life function selection submodule, configured to select a corresponding life evolution function according to the degradation trend type in the life assessment model; A life curve generation submodule, configured to substitute the current detection condition information as an input variable into the life evolution function to obtain a life response curve; The life result analysis submodule is used to perform trend analysis on the life response curve to obtain the life assessment result.

[0081] The specific definition of a capacitor life detection device can be found in the definition of a capacitor life detection method described above and will not be repeated here. Each module in the above-described capacitor life detection device can be implemented in whole or in part through software, hardware, or a combination thereof. Each of these modules can be embedded in or independent of a processor in a computer device in hardware form, or stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0082] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 9As shown. The computer device includes a processor, memory, a network interface, and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When executed by the processor, the computer program implements a capacitor life detection method.

[0083] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed: Obtaining the electrical performance response information of the capacitor under multiple excitation conditions; Based on the electrical performance response information, the dynamic response behavior path of the capacitor in the state space is constructed to obtain the evolutionary behavior sequence; Match the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type; Get current detection condition information; The degradation trend type is input into the life assessment model and analyzed in combination with the current detection condition information to obtain the life assessment result.

[0084] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: Obtaining the electrical performance response information of the capacitor under multiple excitation conditions; Based on the electrical performance response information, the dynamic response behavior path of the capacitor in the state space is constructed to obtain the evolutionary behavior sequence; Match the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type; Get current detection condition information; The degradation trend type is input into the life assessment model and analyzed in combination with the current detection condition information to obtain the life assessment result.

[0085] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0086] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0087] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A capacitor life detection method, characterized in that: The capacitor life detection method comprises: Obtaining the electrical performance response information of the capacitor under multiple excitation conditions; Based on the electrical performance response information, constructing a dynamic response behavior path of the capacitor in a state space to obtain an evolutionary behavior sequence; Matching the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type; Get current detection condition information; The degradation trend type is input into the life assessment model and analyzed in combination with the current detection condition information to obtain a life assessment result.

2. A capacitor life detection method according to claim 1, characterized in that: The step of constructing a dynamic response behavior path of the capacitor in a state space based on the electrical performance response information to obtain an evolutionary behavior sequence includes: Preprocessing the electrical performance response information to obtain preprocessed multi-dimensional response parameters; Mapping the preprocessed multidimensional response parameters into the state space to form a state point sequence; The state point sequence is arranged in the order of changes in the incentive conditions, and a state path structure is established with the incentive dimension as an index to generate the evolutionary behavior sequence.

3. A capacitor life detection method according to claim 2, characterized in that: Arranging the state point sequence according to the change order of the incentive conditions, establishing a state path structure with the incentive dimension as an index, and generating the evolutionary behavior sequence includes: Using the frequency value in the excitation dimension as the index axis, sorting the state point sequence in ascending order of frequency to obtain a sorted state point sequence; Based on the sorted state point sequence, a connection relationship between the state points is established to construct the evolution behavior sequence.

4. A capacitor life detection method according to claim 1, characterized in that: Matching the evolution behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type includes: Calculating the similarity between the evolutionary behavior sequence and the aging behavior path in the aging trajectory library to obtain a similarity score; Based on the similarity score, a matching aging behavior path is confirmed, and according to the matching aging behavior path, the corresponding degradation trend type is identified.

5. A capacitor life detection method according to claim 4, characterized in that: The calculating the similarity between the evolution behavior sequence and the aging behavior path in the aging trajectory library to obtain a similarity score includes: Performing dimension normalization processing on the corresponding state nodes in the evolutionary behavior sequence and the aging behavior path to obtain a normalized state point sequence pair; Based on the normalized state point sequence pairs, calculating the Euclidean distance between corresponding state nodes to obtain a distance set; The distance set is weightedly accumulated to obtain the similarity score.

6. The capacitor life detection method according to claim 1, characterized in that: The degradation trend type is input into the life assessment model and analyzed in combination with the current detection condition information to obtain a life assessment result, including: By selecting a corresponding life evolution function for the degradation trend type in the life assessment model; Substituting the current detection condition information as an input variable into the life evolution function to obtain a life response curve; Performing trend analysis on the life response curve to obtain the life evaluation result.

7. A capacitor life detection method according to claim 6, characterized in that: The performing trend analysis on the life response curve to obtain the life assessment result includes: Fitting the life response curve based on the least square method to generate a life change fitting function; Solving the intersection of the life change fitting function using a preset life determination threshold to obtain the horizontal coordinate of the intersection of the preset life determination threshold and the life change fitting function; The time value corresponding to the abscissa of the intersection is used as the life evaluation result of the capacitor.

8. A capacitor life detection device, characterized in that: The capacitor life detection device comprises: Electrical performance data acquisition module, used to obtain the electrical performance response information of the capacitor under multiple excitation conditions; A behavior trajectory modeling module is used to construct a dynamic response behavior path of the capacitor in a state space based on the electrical performance response information to obtain an evolutionary behavior sequence; a trend identification module, configured to match the evolutionary behavior sequence with the aging behavior path in the aging trajectory library to identify the corresponding degradation trend type; Environmental parameter acquisition module, used to obtain current detection condition information; The life assessment analysis module is used to input the degradation trend type into the life assessment model, and analyze it in combination with the current detection condition information to obtain a life assessment result.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the capacitor life detection method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the capacitor life detection method according to any one of claims 1 to 7 are implemented.

Images