A method and system for capacity fade treatment of a metallized film capacitor

Abstract

The application belongs to the technical field of electronic equipment fault processing, and discloses a kind of metallized film capacitor capacity attenuation processing method and system, comprising: first, the historical data of integrated circuit is acquired, and the corresponding relationship library of running standard vector and component data is established;Then, a circuit simulation model is established, and a capacitance characteristic deviation vector is calculated;When detecting real-time running parameter anomaly, a real-time running vector is constructed and similar historical state is matched from the relationship library, to obtain corresponding component data;The simulation running vector is obtained using the simulation model, and is corrected using the capacitance characteristic deviation vector;The correlation between the corrected running vector and the real-time running vector is analyzed;If there is correlation, the capacitor component data is extracted and an analysis report is output;If there is no correlation, it is determined as a non-capacitor problem and a corresponding report is generated. The fault diagnosis efficiency and accuracy are improved, and the rapid positioning and maintenance of the faulty capacitor are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electronic equipment fault processing, more particularly, the present application relates to a metallized film capacitor capacity attenuation processing method and system. BACKGROUND

[0002] With the rapid development of electronic technology, integrated circuits play an increasingly important role in modern electronic devices. As one of the key components in integrated circuits, capacitors play an indispensable role in signal filtering, energy storage, decoupling and voltage stabilization. However, due to various factors such as temperature changes, voltage stress and aging, the performance of capacitors may decay over time, causing integrated circuits to malfunction or fail. This decay not only affects the normal operation of the circuit, but also may lead to performance degradation or failure of the entire system.

[0003] When integrated circuits malfunction, traditional fault diagnosis methods often require technicians to detect each component in the circuit one by one. This method not only consumes time and effort, but also makes it difficult to accurately locate the problem in complex integrated circuits. Especially for passive components such as capacitors, their performance decay is often gradual, making diagnosis even more difficult. In order to improve the efficiency of fault diagnosis, some existing technologies use circuit simulation software to simulate possible problem scenarios to identify and locate faulty components. Although this method improves the efficiency of diagnosis to some extent, the simulation results of the simulation software have some deviation from the actual situation. This makes the fault diagnosis method based on simulation inaccurate in dealing with capacitor-related problems, thereby leading to low maintenance efficiency. SUMMARY

[0004] In order to overcome the problem of low maintenance efficiency of the prior art, the present application proposes a metallized film capacitor capacity attenuation processing method and system to solve the above problems.

[0005] The present application provides the following technical solutions:

[0006] A metallized film capacitor capacity attenuation processing method, comprising:

[0007] Obtain the historical data of the integrated circuit, the historical data including operating parameters, component data, and establish a running standard vector and component data correspondence relationship library according to the historical data;

[0008] Establish a circuit simulation model according to the actual structure of the integrated circuit, and calculate a capacitor characteristic deviation vector based on the output data of the historical data and the circuit simulation model;

[0009] Collect and monitor the real-time operating parameters of the integrated circuit. When there are operating parameters in the real-time operating parameters that exceed the preset threshold, construct a real-time operating vector based on the real-time operating parameters.

[0010] Based on the real-time running vector, the corresponding component data is obtained from the mapping database between the running standard vector and the component data;

[0011] Based on the obtained component data and circuit simulation model, the simulation running vector is obtained; the capacitor characteristic deviation vector is used to correct the simulation running vector to obtain the corrected running vector.

[0012] Analyze whether there is a correlation between the corrected running vector and the real-time running vector;

[0013] If a correlation exists, extract the component data of type capacitor and output a capacitor analysis report based on the extracted component data.

[0014] If no correlation is found, the problem is determined to be non-capacitor-related and a corresponding report is generated.

[0015] Preferably, the component data includes a serial number, type, and attributes; the operating parameters include output voltage, efficiency, power factor, and total harmonic distortion.

[0016] Preferably, the step of establishing a database of correspondences between running vectors and component data based on historical data includes:

[0017] Extract component data of type capacitor from historical data;

[0018] Based on the extracted component data, historical data that are of the same type (capacitor) and have the same number and attributes are grouped together.

[0019] For each set of historical data, extract the corresponding operating parameters for each historical data point, and normalize the extracted operating parameters to obtain four normalized values. Use the four normalized values ​​to construct the corresponding standard vector.

[0020] For each set of historical data, for all standard vectors, one standard vector is selected as the running standard vector based on the distance between the standard vectors, and the correspondence between the component data in the set of historical data and the running standard vector is established.

[0021] The correspondence between component data and operating standard vectors in each set of historical data is integrated to form a database of correspondence between operating standard vectors and component data.

[0022] Preferably, the step of obtaining one of the standard vectors as the running standard vector based on the distance between the standard vectors includes:

[0023] For each set of historical data corresponding to all standard vectors, calculate the sum of its Euclidean distances to all other standard vectors as the vector distance;

[0024] Set a distance threshold, and mark standard vectors whose vector distance exceeds the distance threshold as abnormal vectors and exclude them;

[0025] For the remaining standard vectors, recalculate the vector distances and select the standard vector with the smallest vector distance as the running standard vector.

[0026] Preferably, the step of calculating the capacitance characteristic deviation vector based on historical data and the output data of the circuit simulation model includes:

[0027] Select any baseline historical data point from the historical data, and obtain the baseline deviation vector based on the baseline historical data and the circuit simulation model; the baseline historical data is the historical data in which all component attributes are at their rated values ​​in the component data;

[0028] Select x historical capacitance difference data from the historical data, and obtain the total deviation vector based on the historical capacitance difference data and the circuit simulation model; where the historical capacitance difference data are: historical data in the component data where the capacitance attribute deviates from the rated value while the other component attributes remain at the rated value, where x≥2;

[0029] The capacitance characteristic deviation vector is obtained by subtracting the reference deviation vector from the total deviation vector.

[0030] Preferably, the step of obtaining the benchmark deviation vector based on the benchmark historical data and the circuit simulation model includes:

[0031] Extract the operating parameters from the historical benchmark data and construct an actual benchmark operating vector from the extracted operating parameters;

[0032] In the circuit simulation model, the properties of all components are set to their rated values, and the simulation is performed to obtain the operating parameters. The operating parameters obtained from the simulation are then used to construct a simulation reference operating vector.

[0033] The benchmark deviation vector is calculated using the following formula:

[0034] B = (RO - SO) / SO

[0035] Where B represents the reference deviation vector, RO represents the actual reference running vector, SO represents the simulated reference running vector, and . / represents the element-wise division operation of the vectors;

[0036] The step of obtaining the total deviation vector based on historical capacitance difference data and circuit simulation model includes:

[0037] Extract the operating parameters of each of the x historical data of capacitance difference, construct an actual capacitance deviation operating vector for each extracted operating parameter, and form an actual capacitance deviation operating vector for all actual capacitance deviation operating vectors.

[0038] In the circuit simulation model, the properties of non-capacitor components are set to their rated values, while the properties of capacitor components are set according to the properties in x historical data of capacitor differences. X simulations are performed to obtain x sets of simulation running parameters. Each set of simulation running parameters is used to construct a simulated capacitor deviation running vector, and all simulated capacitor deviation running vectors constitute a simulated capacitor set.

[0039] Based on the actual capacitor set and the simulated capacitor set, the capacitor deviation vector set is obtained using the following formula:

[0040] D i =(Q i -F i ). / F i ,

[0041] In the formula, D i Q represents the i-th element in the set of capacitance deviation vectors. i F represents the i-th element in the actual set of capacitors. i This represents the i-th element in the set of analog capacitors; i represents the index, 1≤i≤x;

[0042] Calculate the total deviation vector using the following formula:

[0043]

[0044] In the formula, Z represents the total deviation vector, and W i W represents the i-th weight value; i Obtain it using the following formula:

[0045]

[0046] In the formula, ||·|| represents the norm operation, and D j This represents the j-th element in the set of capacitance deviation vectors.

[0047] Preferably, the step of obtaining the corresponding component data from the correspondence database between the running standard vector and component data based on the real-time running vector includes:

[0048] Calculate the cosine similarity between the real-time running vector, the running standard vector, and each running standard vector in the component data correspondence database;

[0049] Based on the calculated cosine similarity, select the standard vector with the highest cosine similarity.

[0050] Extract the component data corresponding to the standard vector with the highest cosine similarity as the final corresponding component data.

[0051] Preferably, obtaining the simulation run vector based on the acquired component data and circuit simulation model includes:

[0052] In the circuit simulation model, the properties of non-capacitor components are set to their rated values, while the properties of capacitor components are set according to the properties in the obtained component data. Simulation is performed to obtain the running parameters, and the obtained running parameters are used to construct a simulation running vector.

[0053] The method of using the capacitance characteristic deviation vector to correct the simulation running vector to obtain the corrected running vector includes:

[0054] Each component in the capacitance characteristic deviation vector is multiplied by the corresponding component in the simulation running vector to obtain the preliminary correction value;

[0055] The initial correction value is added to the corresponding component in the simulation running vector to obtain the corrected running vector.

[0056] Preferably, whether there is a correlation between the analysis and correction running vector and the real-time running vector includes:

[0057] Calculate the normalized Euclidean distance between the corrected running vector and the real-time running vector, and use the reciprocal of the calculated normalized Euclidean distance as the similarity score; compare the similarity score with a preset similarity threshold. If the similarity score is greater than or equal to the preset similarity threshold, it is determined that there is a correlation; otherwise, it is determined that there is no correlation.

[0058] The present invention also provides a system for treating the capacitance decay of metallized film capacitors, for implementing a method for treating the capacitance decay of metallized film capacitors, comprising:

[0059] The relational database module is used to acquire historical data of integrated circuits, including operating parameters and component data. Based on the historical data, a relational database is established between the operating standard vector and the component data.

[0060] The simulation deviation calculation module is used to establish a circuit simulation model based on the actual structure of the integrated circuit, and calculate the capacitance characteristic deviation vector based on historical data and the output data of the circuit simulation model.

[0061] The real-time monitoring module is used to collect and monitor the real-time operating parameters of the integrated circuit. When there are operating parameters in the real-time operating parameters that exceed the preset threshold, a real-time operating vector is constructed based on the real-time operating parameters.

[0062] The data matching module is used to obtain the corresponding component data from the correspondence database between the running standard vector and the component data based on the real-time running vector;

[0063] The simulation calibration module is used to obtain the simulation running vector based on the acquired component data and circuit simulation model; and to calibrate the simulation running vector using the capacitor characteristic deviation vector to obtain the calibrated running vector.

[0064] The correlation analysis module is used to analyze whether there is a correlation between the calibration running vector and the real-time running vector;

[0065] If a correlation exists, extract the component data of type capacitor and output a capacitor analysis report based on the extracted component data.

[0066] If no correlation is found, the problem is determined to be non-capacitor-related and a corresponding report is generated.

[0067] This invention provides a method and system for treating capacitance decay in metallized thin-film capacitors, comprising the following:

[0068] Beneficial effects:

[0069] By extracting relevant information from historical data, the system can quickly match the most similar historical state when an anomaly in real-time operating parameters is detected. By combining historical data and a circuit simulation model, a capacitor characteristic deviation vector is calculated, reflecting the impact of capacitors on circuit performance under different conditions. Using this characteristic deviation vector to correct simulation results effectively reduces the error between simulation and actual conditions, improving the accuracy of simulation-based fault diagnosis methods. By comparing the correlation between the corrected operating vector and the real-time operating vector, the system can intelligently determine whether the current anomaly is caused by a capacitor. When the problem is determined to be caused by a capacitor, the specific faulty capacitor can be quickly located based on the component number information in the component data, shortening fault location time and improving maintenance efficiency. Attached Figure Description

[0070] Figure 1 This is a schematic flowchart of a method for treating capacitance decay in a metallized thin-film capacitor according to the present invention.

[0071] Figure 2 This is a schematic diagram of a metallized thin-film capacitor capacitance decay processing system according to the present invention. Detailed Implementation

[0072] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0073] Example 1

[0074] Please see Figure 1 In this embodiment, a method for treating capacitance decay in a metallized thin-film capacitor includes:

[0075] S1. Obtain historical data of integrated circuits, including operating parameters and component data, and establish a database of correspondence between operating standard vectors and component data based on the historical data;

[0076] The component data includes the component number, type, and attributes; the operating parameters include output voltage, efficiency, power factor, and total harmonic distortion.

[0077] The process of establishing a database of correspondences between operating vectors and component data based on historical data includes:

[0078] Extract component data of type capacitor from historical data;

[0079] Based on the extracted component data, historical data that are of the same type (capacitor) and have the same number and attributes are grouped together.

[0080] For each set of historical data, extract the corresponding operating parameters for each historical data point, and normalize the extracted operating parameters to obtain four normalized values. Use the four normalized values ​​to construct the corresponding standard vector.

[0081] For each set of historical data, for all standard vectors, one standard vector is selected as the running standard vector based on the distance between the standard vectors, and the correspondence between the component data in the set of historical data and the running standard vector is established.

[0082] The correspondence between component data and operating standard vectors in each set of historical data is integrated to form a database of correspondence between operating standard vectors and component data.

[0083] The step of obtaining one of the standard vectors as the running standard vector based on the distance between the standard vectors includes:

[0084] For each set of historical data corresponding to all standard vectors, calculate the sum of its Euclidean distances to all other standard vectors as the vector distance;

[0085] Set a distance threshold, and mark standard vectors whose vector distance exceeds the distance threshold as abnormal vectors and exclude them;

[0086] For the remaining standard vectors, recalculate the vector distances and select the standard vector with the smallest vector distance as the running standard vector.

[0087] In this embodiment, historical data is first acquired. This historical data can be obtained from an experimental environment or from a real operating environment. This data includes operating parameters (output voltage, efficiency, power factor, and total harmonic distortion) and component data (the number, type, and corresponding attributes of each component). Component types include capacitors, resistors, inductors, etc., and the attributes correspond to the component types, such as the capacitance value of a capacitor, the resistance value of a resistor, and the inductance value of an inductor.

[0088] Next, extract all component data of type capacitor from historical data. Since the number of each component in an integrated circuit is usually greater than one, such as multiple capacitors, historical data typically contains information on multiple capacitors. Each component has its own unique number, such as dr001, dr002, etc. Historical data of type capacitor with identical numbers and attributes are grouped together. For example, if multiple historical data entries show capacitor 001 with a capacitance value of 0.5μF, capacitor 002 with a capacitance value of 0.3μF, and all other capacitors with identical numbers and attributes, then these historical data entries will be grouped together.

[0089] For each set of historical data, the corresponding operating parameters are extracted, and the extracted operating parameters are normalized to obtain four normalized values. These four normalized values ​​are then used to construct the corresponding standard vector.

[0090] Then, for each set of historical data and all corresponding standard vectors, calculate the sum of the Euclidean distances of each vector to all other vectors, and use this sum as the vector distance for that vector. Set an appropriate distance threshold, and mark standard vectors whose vector distance exceeds the threshold as outliers and exclude them. For the remaining standard vectors, recalculate the vector distances, and select the standard vector with the smallest vector distance as the running standard vector for that set.

[0091] Next, establish the correspondence between the component data in this group of historical data and the selected operating standard vectors. Repeat the above process for all groups, and finally integrate the correspondence between the component data and the operating standard vectors in each group of historical data to form a complete database of correspondence between operating standard vectors and component data.

[0092] S2. Establish a circuit simulation model based on the actual structure of the integrated circuit, and calculate the capacitance characteristic deviation vector based on historical data and the output data of the circuit simulation model.

[0093] The step of calculating the capacitance characteristic deviation vector based on historical data and circuit simulation model output data includes:

[0094] Select any baseline historical data point from the historical data, and obtain the baseline deviation vector based on the baseline historical data and the circuit simulation model; the baseline historical data is the historical data in which all component attributes are at their rated values ​​in the component data;

[0095] Select x historical capacitance difference data from the historical data, and obtain the total deviation vector based on the historical capacitance difference data and the circuit simulation model; where the historical capacitance difference data are: historical data in the component data where the capacitance attribute deviates from the rated value while the other component attributes remain at the rated value, where x≥2;

[0096] The capacitance characteristic deviation vector is obtained by subtracting the reference deviation vector from the total deviation vector.

[0097] The step of obtaining the benchmark deviation vector based on historical benchmark data and circuit simulation model includes:

[0098] Extract the operating parameters from the historical benchmark data and construct an actual benchmark operating vector from the extracted operating parameters;

[0099] In the circuit simulation model, the properties of all components are set to their rated values, and the simulation is performed to obtain the operating parameters. The operating parameters obtained from the simulation are then used to construct a simulation reference operating vector.

[0100] The benchmark deviation vector is calculated using the following formula:

[0101] B = (RO - SO) / SO

[0102] Where B represents the reference deviation vector, RO represents the actual reference running vector, SO represents the simulated reference running vector, and . / represents the element-wise division operation of the vectors;

[0103] The step of obtaining the total deviation vector based on historical capacitance difference data and circuit simulation model includes:

[0104] Extract the operating parameters of each of the x historical data of capacitance difference, construct an actual capacitance deviation operating vector for each extracted operating parameter, and form an actual capacitance deviation operating vector for all actual capacitance deviation operating vectors.

[0105] In the circuit simulation model, the properties of non-capacitor components are set to their rated values, while the properties of capacitor components are set according to the properties in x historical data of capacitor differences. X simulations are performed to obtain x sets of simulation running parameters. Each set of simulation running parameters is used to construct a simulated capacitor deviation running vector, and all simulated capacitor deviation running vectors constitute a simulated capacitor set.

[0106] Based on the actual capacitor set and the simulated capacitor set, the capacitor deviation vector set is obtained using the following formula:

[0107] D i =(Q i -F i ). / F i ,

[0108] In the formula, D i Q represents the i-th element in the set of capacitance deviation vectors. i F represents the i-th element in the actual set of capacitors. i This represents the i-th element in the set of analog capacitors; i represents the index, 1≤i≤x;

[0109] Calculate the total deviation vector using the following formula:

[0110]

[0111] In the formula, Z represents the total deviation vector, and W i W represents the i-th weight value; i Obtain it using the following formula:

[0112]

[0113] In the formula, ||·|| represents the norm operation, and D j This represents the j-th element in the set of capacitance deviation vectors.

[0114] In this embodiment, a circuit simulation model is first established using circuit simulation software based on the actual structure of the integrated circuit. Next, the capacitance characteristic deviation vector is calculated based on historical data and the output data of the circuit simulation model. The calculation process is as follows:

[0115] First, select a baseline historical data set where all component properties are at their rated values. For example, if a circuit has a capacitor with a rated capacitance of 100μF, then in this baseline historical data set, the capacitance of this capacitor will be 100μF. Next, calculate the baseline deviation vector. Extract the actual operating parameters (output voltage, efficiency, power factor, and total harmonic distortion) from the baseline historical data to construct the actual baseline operating vector. Set all components to their rated values ​​in the circuit simulation model and perform a simulation to obtain the simulated baseline operating vector. Divide the actual baseline operating vector by the simulated baseline operating vector to obtain the baseline deviation vector. This vector reflects the inherent error between the simulation model and the actual circuit.

[0116] Next, n historical capacitance deviation data points are selected from the historical data. In these data points, the capacitor's properties deviate from their rated values, while the properties of other components remain at their rated values. For example, in one data point, a 100μF capacitor's actual value changes to 95μF, while other components remain at their rated values. For each historical capacitance deviation data point, its actual operating parameters are extracted to construct an actual capacitance deviation operating vector. All these vectors constitute the actual capacitance set. The corresponding capacitance values ​​are set in the simulation model, and n simulations are performed to obtain the simulated capacitance deviation operating vector, which constitutes the simulated capacitance set.

[0117] Divide the actual set of capacitors by the simulated set of capacitors to obtain a set of capacitor deviation vectors. Calculate the norm of each capacitor deviation vector, and use the reciprocal of this norm, after normalization, as the weight. Sum all the capacitor deviation vectors using weighted averages to obtain the total deviation vector.

[0118] Finally, the reference deviation vector is subtracted from the total deviation vector to obtain the capacitor characteristic deviation vector. This vector reflects the specific error caused by the capacitance value deviating from the rated value. The capacitor characteristic deviation vector is used to correct the error between the simulation results and the actual results when the capacitor is not rated. Applying this deviation vector improves the accuracy of the simulation of the actual circuit behavior and provides a reliable data foundation for subsequent analysis of capacitor problems in integrated circuits.

[0119] S3. Collect and monitor the real-time operating parameters of the integrated circuit. When there are operating parameters in the real-time operating parameters that exceed the preset threshold, construct a real-time operating vector based on the real-time operating parameters.

[0120] S4. Obtain the corresponding component data from the correspondence database between the running standard vector and component data based on the real-time running vector;

[0121] The step of obtaining the corresponding component data from the correspondence database between the running standard vector and component data based on the real-time running vector includes:

[0122] Calculate the cosine similarity between the real-time running vector, the running standard vector, and each running standard vector in the component data correspondence database;

[0123] Based on the calculated cosine similarity, select the standard vector with the highest cosine similarity.

[0124] Extract the component data corresponding to the standard vector with the highest cosine similarity as the final corresponding component data.

[0125] In this embodiment, the real-time operating parameters of the integrated circuit are continuously collected and monitored. These parameters include output voltage, efficiency, power factor, and total harmonic distortion. A normal operating threshold range for each parameter is preset. When any real-time operating parameter is detected to exceed its preset threshold, it indicates a potential fault in the integrated circuit, and therefore immediately triggers the following processing flow:

[0126] First, the current operating parameters are combined to construct a real-time operating vector. Next, this real-time operating vector is used to retrieve matching component data from a previously established database of mappings between standard operating vectors and component data. This process employs cosine similarity calculation; after calculation, the standard operating vector with the highest cosine similarity is selected. This vector represents the historical operating state most similar to the current real-time state. Then, the component data associated with this most similar standard operating vector is extracted; this represents the most likely component data for the current state.

[0127] S5. Based on the obtained component data and circuit simulation model, obtain the simulation running vector; use the capacitor characteristic deviation vector to correct the simulation running vector to obtain the corrected running vector;

[0128] The process of obtaining the simulation run vector based on the acquired component data and circuit simulation model includes:

[0129] In the circuit simulation model, the properties of non-capacitor components are set to their rated values, while the properties of capacitor components are set according to the properties in the obtained component data. Simulation is performed to obtain the running parameters, and the obtained running parameters are used to construct a simulation running vector.

[0130] The method of using the capacitance characteristic deviation vector to correct the simulation running vector to obtain the corrected running vector includes:

[0131] Each component in the capacitance characteristic deviation vector is multiplied by the corresponding component in the simulation running vector to obtain the preliminary correction value;

[0132] The initial correction value is added to the corresponding component in the simulation running vector to obtain the corrected running vector.

[0133] In this embodiment, the simulation run vector is generated using previously acquired component data and the established circuit simulation model. The purpose of this process is to simulate a circuit state where a capacitor may malfunction while other components remain normal. The specific process is as follows:

[0134] First, component parameters are set in the circuit simulation model. For non-capacitor components, their properties are set to their rated values ​​to simulate normal operation. For capacitor components, their properties are precisely set based on previously acquired component data; these properties may deviate from their rated values, reflecting potential abnormal states. After setting, the simulation is run, yielding a set of operating parameters. These simulation parameters are then used to construct a simulation operating vector. This vector reflects the theoretical operating state of the circuit under conditions where capacitors may malfunction while other components function normally. Since simulation results often deviate from actual conditions, the simulation operating vector is corrected using a previously calculated capacitor characteristic deviation vector to improve simulation accuracy.

[0135] The correction process is as follows: First, each component of the capacitor characteristic deviation vector is multiplied by its corresponding component in the simulation running vector to obtain a preliminary correction value. This step considers the impact of capacitor characteristic changes on various operating parameters. Next, these preliminary correction values ​​are added to their corresponding components in the simulation running vector to obtain the final corrected running vector. This corrected running vector integrates the capacitor characteristic deviations observed in simulation results and historical data, thus more accurately reflecting the actual circuit's operating state under abnormal capacitor conditions.

[0136] S6. Analyze whether there is a correlation between the corrected running vector and the real-time running vector;

[0137] If a correlation exists, extract the component data of type capacitor and output a capacitor analysis report based on the extracted component data.

[0138] If no correlation is found, the problem is determined to be non-capacitor-related and a corresponding report is generated.

[0139] The analysis of whether there is a correlation between the corrected running vector and the real-time running vector includes:

[0140] Calculate the normalized Euclidean distance between the corrected running vector and the real-time running vector, and use the reciprocal of the calculated normalized Euclidean distance as the similarity score; compare the similarity score with a preset similarity threshold. If the similarity score is greater than or equal to the preset similarity threshold, it is determined that there is a correlation; otherwise, it is determined that there is no correlation.

[0141] In this embodiment, the correlation between the corrected operating vector and the real-time operating vector is further analyzed to determine whether the current abnormal operating parameters are caused by the capacitor. First, the normalized Euclidean distance between the corrected operating vector and the real-time operating vector is calculated, and the reciprocal of this distance is used as the similarity score. This method can effectively quantify the similarity between the two vectors. The obtained similarity score is compared with a preset similarity threshold. If the similarity score is greater than or equal to the preset threshold, the two vectors are determined to be correlated; otherwise, they are determined not to be correlated.

[0142] When a correlation is determined, it indicates that the observed abnormal operating parameters are likely due to a capacitor problem. In this case, capacitor data from previous simulations are extracted, as this data usually reflects the actual state of the capacitors in the current circuit well. A detailed capacitor analysis report is generated based on this data. The report includes the current estimated capacitance value, the rated capacitance value, and the difference between the two for each relevant capacitor. If the difference between the actual value and the rated value of a capacitor exceeds a predetermined threshold, the report will recommend replacing the capacitor or conducting more detailed testing. Because each capacitor has a unique number, the report can accurately pinpoint the specific capacitor requiring attention, greatly improving the efficiency of maintenance and repair. For example, the report might state: Capacitor numbered dr001 has a current estimated capacitance value of 95μF, a rated capacitance value of 100μF, and a deviation rate of 5%. Further testing is recommended. Or, capacitor numbered dr002 has a current estimated capacitance value of 80μF, a rated capacitance value of 100μF, and a deviation rate of 20%. Replacement of this capacitor is recommended.

[0143] Conversely, if it is determined that there is no correlation between the corrected operating vector and the real-time operating vector, this indicates that the currently observed abnormal operating parameters may not be caused by a capacitor problem. In this case, a different report will be generated, indicating that the problem may lie with other types of components.

[0144] This method allows maintenance personnel to quickly and accurately focus on the components most likely to cause problems, whether capacitors or other types of components. This significantly improves the efficiency of fault diagnosis and repair, reduces unnecessary inspection time, and enhances the accuracy of maintenance.

[0145] Example 2

[0146] Please see Figure 2 This invention provides a system for treating the capacitance decay of metallized film capacitors, used to implement a method for treating the capacitance decay of metallized film capacitors, comprising:

[0147] The relational database module is used to acquire historical data of integrated circuits, including operating parameters and component data. Based on the historical data, a relational database is established between the operating standard vector and the component data.

[0148] The simulation deviation calculation module is used to establish a circuit simulation model based on the actual structure of the integrated circuit, and calculate the capacitance characteristic deviation vector based on historical data and the output data of the circuit simulation model.

[0149] The real-time monitoring module is used to collect and monitor the real-time operating parameters of the integrated circuit. When there are operating parameters in the real-time operating parameters that exceed the preset threshold, a real-time operating vector is constructed based on the real-time operating parameters.

[0150] The data matching module is used to obtain the corresponding component data from the correspondence database between the running standard vector and the component data based on the real-time running vector;

[0151] The simulation calibration module is used to obtain the simulation running vector based on the acquired component data and circuit simulation model; and to calibrate the simulation running vector using the capacitor characteristic deviation vector to obtain the calibrated running vector.

[0152] The correlation analysis module is used to analyze whether there is a correlation between the calibration running vector and the real-time running vector;

[0153] If a correlation exists, extract the component data of type capacitor and output a capacitor analysis report based on the extracted component data.

[0154] If no correlation is found, the problem is determined to be non-capacitor-related and a corresponding report is generated.

[0155] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0156] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0157] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for treating capacitance decay in metallized thin-film capacitors, characterized in that, include: Acquire historical data of integrated circuits, including operating parameters and component data, and establish a database of correspondence between operating standard vectors and component data based on the historical data; A circuit simulation model is established based on the actual structure of the integrated circuit, and the capacitance characteristic deviation vector is calculated based on historical data and the output data of the circuit simulation model. Collect and monitor the real-time operating parameters of the integrated circuit. When there are operating parameters in the real-time operating parameters that exceed the preset threshold, construct a real-time operating vector based on the real-time operating parameters. Based on the real-time running vector, the corresponding component data is obtained from the mapping database between the running standard vector and the component data; Based on the obtained component data and circuit simulation model, the simulation running vector is obtained; the capacitor characteristic deviation vector is used to correct the simulation running vector to obtain the corrected running vector. Analyze whether there is a correlation between the corrected running vector and the real-time running vector; If a correlation exists, extract the component data of type capacitor and output a capacitor analysis report based on the extracted component data. If no correlation is found, the problem is determined to be non-capacitor-related and a corresponding report is generated.

2. The method for treating capacitance decay of a metallized thin-film capacitor according to claim 1, characterized in that, The component data includes the component number, type, and attributes; the operating parameters include output voltage, efficiency, power factor, and total harmonic distortion.

3. The method for treating capacitance decay in a metallized thin-film capacitor according to claim 2, characterized in that, The process of establishing a database of correspondences between operating vectors and component data based on historical data includes: Extract component data of type capacitor from historical data; Based on the extracted component data, historical data that are of the same type (capacitor) and have the same number and attributes are grouped together. For each set of historical data, extract the corresponding operating parameters for each historical data point, and normalize the extracted operating parameters to obtain four normalized values. Use the four normalized values ​​to construct the corresponding standard vector. For each set of historical data, for all standard vectors, one standard vector is selected as the running standard vector based on the distance between the standard vectors, and the correspondence between the component data in the set of historical data and the running standard vector is established. The correspondence between component data and operating standard vectors in each set of historical data is integrated to form a database of correspondence between operating standard vectors and component data.

4. The method for treating capacitance decay in a metallized thin-film capacitor according to claim 3, characterized in that, The step of obtaining one of the standard vectors as the running standard vector based on the distance between the standard vectors includes: For each set of historical data corresponding to all standard vectors, calculate the sum of its Euclidean distances to all other standard vectors as the vector distance; Set a distance threshold, and mark standard vectors whose vector distance exceeds the distance threshold as abnormal vectors and exclude them; For the remaining standard vectors, recalculate the vector distances and select the standard vector with the smallest vector distance as the running standard vector.

5. The method for treating capacitance decay in a metallized thin-film capacitor according to claim 1, characterized in that, The step of calculating the capacitance characteristic deviation vector based on historical data and circuit simulation model output data includes: Select any baseline historical data point from the historical data, and obtain the baseline deviation vector based on the baseline historical data and the circuit simulation model; the baseline historical data is the historical data in which all component attributes are at their rated values ​​in the component data. Select x historical capacitance difference data from the historical data, and obtain the total deviation vector based on the historical capacitance difference data and the circuit simulation model; where the historical capacitance difference data are: historical data in the component data where the capacitance attribute deviates from the rated value while the other component attributes remain at the rated value, where x≥2; The capacitance characteristic deviation vector is obtained by subtracting the reference deviation vector from the total deviation vector.

6. The method for treating capacitance decay of a metallized thin-film capacitor according to claim 5, characterized in that, The step of obtaining the benchmark deviation vector based on historical benchmark data and circuit simulation model includes: Extract the operating parameters from the historical benchmark data and construct an actual benchmark operating vector from the extracted operating parameters; In the circuit simulation model, the properties of all components are set to their rated values, and the simulation is performed to obtain the operating parameters. The operating parameters obtained from the simulation are then used to construct a simulation reference operating vector. The benchmark deviation vector is calculated using the following formula: B = (RO - SO) / SO Where B represents the reference deviation vector, RO represents the actual reference running vector, SO represents the simulated reference running vector, and . / represents the element-wise division operation of the vectors; The step of obtaining the total deviation vector based on historical capacitance difference data and circuit simulation model includes: Extract the operating parameters of each of the x historical data of capacitance difference, construct an actual capacitance deviation operating vector for each extracted operating parameter, and form an actual capacitance deviation operating vector for all actual capacitance deviation operating vectors. In the circuit simulation model, the properties of non-capacitor components are set to their rated values, while the properties of capacitor components are set according to the properties in x historical data of capacitor differences. X simulations are performed to obtain x sets of simulation running parameters. Each set of simulation running parameters is used to construct a simulated capacitor deviation running vector, and all simulated capacitor deviation running vectors constitute a simulated capacitor set. Based on the actual capacitor set and the simulated capacitor set, the capacitor deviation vector set is obtained using the following formula: D i =(Q i -F i ). / F i , In the formula, D i Q represents the i-th element in the set of capacitance deviation vectors. i F represents the i-th element in the actual set of capacitors. i This represents the i-th element in the set of analog capacitors; i represents the index, 1≤i≤x; Calculate the total deviation vector using the following formula: In the formula, Z represents the total deviation vector, and W i W represents the i-th weight value; i Obtain it using the following formula: In the formula, ||·|| represents the norm operation, and D j This represents the j-th element in the set of capacitance deviation vectors.

7. The method for treating capacitance decay of a metallized thin-film capacitor according to claim 6, characterized in that, The step of obtaining the corresponding component data from the correspondence database between the running standard vector and component data based on the real-time running vector includes: Calculate the cosine similarity between the real-time running vector, the running standard vector, and each running standard vector in the component data correspondence database; Based on the calculated cosine similarity, select the standard vector with the highest cosine similarity. Extract the component data corresponding to the standard vector with the highest cosine similarity as the final corresponding component data.

8. The method for treating capacitance decay of a metallized thin-film capacitor according to claim 7, characterized in that, The process of obtaining the simulation run vector based on the acquired component data and circuit simulation model includes: In the circuit simulation model, the properties of non-capacitor components are set to their rated values, while the properties of capacitor components are set according to the properties in the obtained component data. Simulation is performed to obtain the running parameters, and the obtained running parameters are used to construct a simulation running vector. The method of using the capacitance characteristic deviation vector to correct the simulation running vector to obtain the corrected running vector includes: Each component in the capacitance characteristic deviation vector is multiplied by the corresponding component in the simulation running vector to obtain the preliminary correction value; The initial correction value is added to the corresponding component in the simulation running vector to obtain the corrected running vector.

9. The method for treating capacitance decay of a metallized thin-film capacitor according to claim 8, characterized in that, The analysis of whether there is a correlation between the corrected running vector and the real-time running vector includes: Calculate the normalized Euclidean distance between the corrected running vector and the real-time running vector, and use the reciprocal of the calculated normalized Euclidean distance as the similarity score; compare the similarity score with a preset similarity threshold. If the similarity score is greater than or equal to the preset similarity threshold, it is determined that there is a correlation; otherwise, it is determined that there is no correlation.

10. A system for treating the capacitance decay of a metallized film capacitor, used to implement the method for treating the capacitance decay of a metallized film capacitor as described in any one of claims 1-9, characterized in that, include: The relational database module is used to acquire historical data of integrated circuits, including operating parameters and component data. Based on the historical data, a relational database is established between the operating standard vector and the component data. The simulation deviation calculation module is used to establish a circuit simulation model based on the actual structure of the integrated circuit, and calculate the capacitance characteristic deviation vector based on historical data and the output data of the circuit simulation model. The real-time monitoring module is used to collect and monitor the real-time operating parameters of the integrated circuit. When there are operating parameters in the real-time operating parameters that exceed the preset threshold, a real-time operating vector is constructed based on the real-time operating parameters. The data matching module is used to obtain the corresponding component data from the correspondence database between the running standard vector and the component data based on the real-time running vector; The simulation calibration module is used to obtain the simulation running vector based on the acquired component data and circuit simulation model; and to calibrate the simulation running vector using the capacitor characteristic deviation vector to obtain the calibrated running vector. The correlation analysis module is used to analyze whether there is a correlation between the calibration running vector and the real-time running vector; If a correlation exists, extract the component data of type capacitor and output a capacitor analysis report based on the extracted component data. If no correlation is found, the problem is determined to be non-capacitor-related and a corresponding report is generated.

Images