CN118091478B - Multichannel combined power supply life monitoring and early warning system based on data analysis - Google Patents
Multichannel combined power supply life monitoring and early warning system based on data analysis Download PDFInfo
- Publication number
- CN118091478B CN118091478B CN202410517918.7A CN202410517918A CN118091478B CN 118091478 B CN118091478 B CN 118091478B CN 202410517918 A CN202410517918 A CN 202410517918A CN 118091478 B CN118091478 B CN 118091478B
- Authority
- CN
- China
- Prior art keywords
- ripple
- stability
- voltage value
- vector
- stability coefficient
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 13
- 238000007405 data analysis Methods 0.000 title claims abstract description 9
- 239000013598 vector Substances 0.000 claims abstract description 106
- 238000007477 logistic regression Methods 0.000 claims description 42
- 238000012216 screening Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 12
- 238000000034 method Methods 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 3
- 230000032683 aging Effects 0.000 abstract description 5
- 238000001514 detection method Methods 0.000 abstract description 5
- 239000000284 extract Substances 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/40—Testing power supplies
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
Abstract
The invention discloses a multi-path combined power life monitoring and early warning system based on data analysis, which belongs to the technical field of power life monitoring, and the invention collects each path of voltage signals, screens out noise voltage values from each path of voltage signals, denoises, removes thermal noise and flicker noise, segments the denoised voltage signals according to peak positions to obtain multi-section voltage sub-signals, extracts a ripple characteristic vector from each section of voltage sub-signals, the waveform shape characteristics of a section of voltage sub-signals are represented, and then ripple stability coefficient vectors are obtained according to the ripple characteristic vectors of the same path, so that the stability of ripple is reflected, the aging condition of each part of devices in the power supply is reflected through the stability of ripple, the life prediction is carried out based on a multi-input life prediction model, the running state of power supply equipment is monitored in high efficiency and in real time, and the missing detection or false alarm of faults is reduced.
Description
Technical Field
The invention relates to the technical field of power supply life monitoring, in particular to a multi-path combined power supply life monitoring and early warning system based on data analysis.
Background
A multiple-channel combined power supply refers to a power supply device capable of providing a plurality of different voltage outputs. In modern electronic systems, different components may require different supply voltages to function properly. For example, a system may require +5v to drive some logic circuits while requiring ±15v or ±12v to drive analog circuits. To meet these needs, multiple combined power supplies are designed to provide a variety of different voltage outputs, thereby ensuring that each component in the system is supplied with the exact and stable power required. Such power supplies are commonly used in complex electronic devices or systems, such as communication devices, industrial control systems, medical devices, etc., that require multiple voltages to ensure proper operation of the different components. The performance and life of the existing power supply equipment are evaluated by detecting the power supply equipment regularly. The method has the advantages of simplicity and easy implementation, but has the defects of low efficiency, failure to monitor the running state of the power supply equipment in real time and possible missed detection or false alarm of faults.
Disclosure of Invention
Aiming at the defects in the prior art, the multi-path combined power supply life monitoring and early warning system based on data analysis solves the problems that the existing method for evaluating the performance and the life of the power supply equipment by periodically detecting the power supply equipment is low in efficiency, cannot monitor the running state of the power supply equipment in real time and causes missed detection or false alarm of faults.
In order to achieve the aim of the invention, the invention adopts the following technical scheme: a multi-path combined power supply life monitoring and early warning system based on data analysis comprises: the device comprises an acquisition unit, a denoising unit, a segmentation unit, a ripple characteristic extraction unit, a ripple stability coefficient extraction unit and a service life early warning unit;
the acquisition unit is used for acquiring voltage signals of each path of the multi-path combined power supply;
The denoising unit is used for screening noise voltage values from each path of voltage signal, and denoising to obtain denoising voltage signals;
The segmentation unit is used for segmenting the denoising voltage signal according to the peak position to obtain a plurality of segments of voltage sub-signals;
the ripple characteristic extraction unit is used for extracting a ripple characteristic vector for each section of voltage sub-signal;
The ripple stability coefficient extraction unit is used for obtaining a ripple stability coefficient vector according to the ripple characteristic vector of the same path;
The life early warning unit is used for predicting the life of the power supply based on the multi-input life prediction model according to the ripple stability coefficient vector, and early warning is carried out when the life of the power supply is lower than a threshold value.
The beneficial effects of the invention are as follows: the invention collects each path of voltage signal, screens out noise voltage value from each path of voltage signal, removes noise treatment, removes thermal noise and flicker noise, segments the noise-removed voltage signal according to peak positions to obtain multi-section voltage sub-signals, extracts a ripple characteristic vector from each section of voltage sub-signals, characterizes waveform shape characteristics of one section of voltage sub-signals, acquires ripple stability coefficient vectors according to the ripple characteristic vector of the same path, thus representing stability of ripple, reflects aging conditions of all parts of devices in a power supply through the stability of ripple, predicts service life based on a multi-input service life prediction model, realizes high-efficiency and real-time monitoring of running state of power supply equipment, and reduces missing detection or false alarm of faults.
In a dc power supply, ideally, the voltage signal is a straight line, but in reality, the voltage signal is doped with noise and some ac components (in non-battery-type electric appliances, power needs to be taken from a power grid, and the power frequency of the power grid is 50Hz, so that the ac components are generated), and the actual voltage signal is not a straight line but a fluctuating line. The invention segments the denoising voltage signal from each peak position, and divides the denoising voltage signal into a plurality of parts, thereby detecting the stability of the ripple characteristic vector on each part, and further realizing the evaluation of the service life of the power supply.
Further, the denoising unit includes: the device comprises a sliding queue, an average gradient calculation module, a noise voltage value screening module and a denoising module;
the sliding queue is N in length and is used for sliding on the voltage signal, filling voltage values in the voltage signal, and the sliding distance is 1 voltage value each time;
the average gradient calculation module is used for calculating an average gradient according to the voltage value in the sliding queue after each sliding of the sliding queue;
the noise voltage value screening module is used for classifying each voltage value into a noise voltage value when the gradient of the voltage value is larger than the average gradient;
the denoising module is used for denoising the noise voltage value to obtain a denoising voltage signal.
Further, the formula for calculating the average gradient is: Wherein d c is an average gradient, v i is the ith voltage value in the sliding queue, v i+1 is the (i+1) th voltage value in the sliding queue, i is a positive integer, and i is an absolute value;
the calculation formula of the gradient of the voltage value is as follows: Wherein d i is the gradient of the ith voltage value in the sliding queue, and v i-1 is the ith-1 voltage value in the sliding queue.
The beneficial effects of the above further scheme are: the method comprises the steps of screening out noise voltage values, carrying out targeted denoising treatment on the noise voltage values, calculating an average gradient in a sliding queue when the noise voltage values are screened out, evaluating the change condition of the voltage values in the sliding queue, and screening out voltage values with abnormal changes as noise voltage values.
Further, the denoising formula is: Wherein, the method comprises the steps of, wherein, For the denoised voltage value, v o is the noise voltage value, v l is the voltage value on the left side of the noise voltage value in the sliding queue, and v r is the voltage value on the right side of the noise voltage value in the sliding queue.
The beneficial effects of the above further scheme are: when the gradient of the voltage value is larger, the denoising intensity of the noise voltage value is larger, and the denoising intensity is adaptively adjusted.
Further, the ripple characteristic vector is: Wherein R j is the jth ripple feature vector, R j,1 is the 1 st element in the jth ripple feature vector, R j,2 is the 2 nd element in the jth ripple feature vector, R j,3 is the 3 rd element in the jth ripple feature vector, R j,4 is the 4 th element in the jth ripple feature vector, the 1 st element in the ripple feature vector is the peak-to-peak value, the 2 nd element in the ripple feature vector is the variance, the 3 rd element in the ripple feature vector is the skewness, the 4 th element in the ripple feature vector is the kurtosis, and j is a positive integer.
The beneficial effects of the above further scheme are: the ripple belongs to a repeatedly existing waveform in the power supply, so that the invention segments the denoising voltage signal according to the peak position to obtain a plurality of segments of voltage sub-signals, extracts the characteristic vector of the ripple in each segment of voltage sub-signal, reduces the data quantity and characterizes the waveform form.
Further, the ripple stability factor vector is: wherein S is a ripple stability coefficient vector, S 1 is the 1 st element in the ripple stability coefficient vector, S 2 is the 2 nd element in the ripple stability coefficient vector, S 3 is the 3 rd element in the ripple stability coefficient vector, S 4 is the 4 th element in the ripple stability coefficient vector, the 1 st element in the ripple stability coefficient vector is a peak-to-peak stability coefficient, the 2 nd element in the ripple stability coefficient vector is a variance stability coefficient, the 3 rd element in the ripple stability coefficient vector is a skewness stability coefficient, and the 4 th element in the ripple stability coefficient vector is a kurtosis stability coefficient;
the calculation formula of each element in the ripple stability coefficient vector is as follows: Wherein s k is the kth element in the ripple stability coefficient vector, the value range of k is 1,2,3,4, r j,k is the kth element in the jth ripple characteristic vector, and M is the number of the ripple characteristic vectors in the same path.
The beneficial effects of the above further scheme are: according to the invention, a plurality of ripple characteristic vectors are extracted from the same path, the ripple condition is measured through each element in the plurality of ripple characteristic vectors, and whether the device in the power supply has aging is measured through the stable condition of the ripple.
Further, the multiple-input life prediction model includes: a plurality of logistic regression layers, an enhancement layer, and a life prediction layer;
The input end of each logistic regression layer is used for inputting ripple stability coefficient vectors of each path, and the output end of each logistic regression layer is used for outputting stability;
The enhancement layer is used for enhancing the stability to obtain enhanced stability;
The life prediction layer is used for predicting the life of the power supply according to the stability after each enhancement.
Further, the expression of each logistic regression layer is: Where h is the stability of the output of the logistic regression layer, ω k is the kth weight in the logistic regression layer, b k is the kth bias in the logistic regression layer, and σ is the Sigmoid function.
The beneficial effects of the above further scheme are: in the invention, a logistic regression layer processes a ripple characteristic vector to realize evaluation of one path of output.
Further, the expression of the enhancement layer is: Wherein h e,m is the stability after the m-th enhancement, h m is the stability of the m-th logistic regression layer output, L is the number of the logistic regression layers, and m is a positive integer.
The beneficial effects of the above further scheme are: the invention distributes the proportion coefficient to the output according to the output h m of each logistic regression layerThe output of each logistic regression layer is adjusted in a self-adaptive mode, and the attention of important features of the multi-input life prediction model is achieved in a self-adaptive mode.
Further, the lifetime prediction layer has the expression: Wherein y is the service life of the power supply, h e,m is the stability after the m-th enhancement, omega m is the m-th weight in the logistic regression layer, b m is the m-th bias in the logistic regression layer, L is the number of the logistic regression layer, and m is a positive integer.
Drawings
FIG. 1 is a system block diagram of a multi-channel combined power supply life monitoring and early warning system based on data analysis.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and all the inventions which make use of the inventive concept are protected by the spirit and scope of the present invention as defined and defined in the appended claims to those skilled in the art.
As shown in fig. 1, a multi-path combined power supply life monitoring and early warning system based on data analysis includes: the device comprises an acquisition unit, a denoising unit, a segmentation unit, a ripple characteristic extraction unit, a ripple stability coefficient extraction unit and a service life early warning unit;
the acquisition unit is used for acquiring voltage signals of each path of the multi-path combined power supply;
The denoising unit is used for screening noise voltage values from each path of voltage signal, and denoising to obtain denoising voltage signals;
The segmentation unit is used for segmenting the denoising voltage signal according to the peak position to obtain a plurality of segments of voltage sub-signals;
the ripple characteristic extraction unit is used for extracting a ripple characteristic vector for each section of voltage sub-signal;
The ripple stability coefficient extraction unit is used for obtaining a ripple stability coefficient vector according to the ripple characteristic vector of the same path;
The life early warning unit is used for predicting the life of the power supply based on the multi-input life prediction model according to the ripple stability coefficient vector, and early warning is carried out when the life of the power supply is lower than a threshold value.
In a dc power supply, ideally, the voltage signal is a straight line, but in reality, the voltage signal is doped with noise and some ac components (in non-battery-type electric appliances, power needs to be taken from a power grid, and the power frequency of the power grid is 50Hz, so that the ac components are generated), and the actual voltage signal is not a straight line but a fluctuating line. The invention segments the denoising voltage signal from each peak position, and divides the denoising voltage signal into a plurality of parts, thereby detecting the stability of the ripple characteristic vector on each part, and further realizing the evaluation of the service life of the power supply.
The denoising unit includes: the device comprises a sliding queue, an average gradient calculation module, a noise voltage value screening module and a denoising module;
the sliding queue is N in length and is used for sliding on the voltage signal, filling voltage values in the voltage signal, and the sliding distance is 1 voltage value each time;
the average gradient calculation module is used for calculating an average gradient according to the voltage value in the sliding queue after each sliding of the sliding queue;
the noise voltage value screening module is used for classifying each voltage value into a noise voltage value when the gradient of the voltage value is larger than the average gradient;
the denoising module is used for denoising the noise voltage value to obtain a denoising voltage signal.
The formula for calculating the average gradient is as follows: Wherein d c is an average gradient, v i is the ith voltage value in the sliding queue, v i+1 is the (i+1) th voltage value in the sliding queue, i is a positive integer, and i is an absolute value;
the calculation formula of the gradient of the voltage value is as follows: Wherein d i is the gradient of the ith voltage value in the sliding queue, and v i-1 is the ith-1 voltage value in the sliding queue.
The method comprises the steps of screening out noise voltage values, carrying out targeted denoising treatment on the noise voltage values, calculating an average gradient in a sliding queue when the noise voltage values are screened out, evaluating the change condition of the voltage values in the sliding queue, and screening out voltage values with abnormal changes as noise voltage values.
The denoising formula is as follows: Wherein, the method comprises the steps of, wherein, For the denoised voltage value, v o is the noise voltage value, v l is the voltage value on the left side of the noise voltage value in the sliding queue, and v r is the voltage value on the right side of the noise voltage value in the sliding queue.
When the gradient of the voltage value is larger, the denoising intensity of the noise voltage value is larger, and the denoising intensity is adaptively adjusted.
The ripple characteristic vector is: Wherein R j is the jth ripple feature vector, R j,1 is the 1 st element in the jth ripple feature vector, R j,2 is the 2 nd element in the jth ripple feature vector, R j,3 is the 3 rd element in the jth ripple feature vector, R j,4 is the 4 th element in the jth ripple feature vector, the 1 st element in the ripple feature vector is the peak-to-peak value, the 2 nd element in the ripple feature vector is the variance, the 3 rd element in the ripple feature vector is the skewness, the 4 th element in the ripple feature vector is the kurtosis, and j is a positive integer.
The ripple belongs to a repeatedly existing waveform in the power supply, so that the invention segments the denoising voltage signal according to the peak position to obtain a plurality of segments of voltage sub-signals, extracts the characteristic vector of the ripple in each segment of voltage sub-signal, reduces the data quantity and characterizes the waveform form.
The ripple stability factor vector is: wherein S is a ripple stability coefficient vector, S 1 is the 1 st element in the ripple stability coefficient vector, S 2 is the 2 nd element in the ripple stability coefficient vector, S 3 is the 3 rd element in the ripple stability coefficient vector, S 4 is the 4 th element in the ripple stability coefficient vector, the 1 st element in the ripple stability coefficient vector is a peak-to-peak stability coefficient, the 2 nd element in the ripple stability coefficient vector is a variance stability coefficient, the 3 rd element in the ripple stability coefficient vector is a skewness stability coefficient, and the 4 th element in the ripple stability coefficient vector is a kurtosis stability coefficient;
the calculation formula of each element in the ripple stability coefficient vector is as follows: Wherein s k is the kth element in the ripple stability coefficient vector, the value range of k is 1,2,3,4, r j,k is the kth element in the jth ripple characteristic vector, and M is the number of the ripple characteristic vectors in the same path.
According to the invention, a plurality of ripple characteristic vectors are extracted from the same path, the ripple condition is measured through each element in the plurality of ripple characteristic vectors, and whether the device in the power supply has aging is measured through the stable condition of the ripple.
The multiple-input life prediction model includes: a plurality of logistic regression layers, an enhancement layer, and a life prediction layer;
The input end of each logistic regression layer is used for inputting ripple stability coefficient vectors of each path, and the output end of each logistic regression layer is used for outputting stability;
The enhancement layer is used for enhancing the stability to obtain enhanced stability;
The life prediction layer is used for predicting the life of the power supply according to the stability after each enhancement.
The expression of each logistic regression layer is: Where h is the stability of the output of the logistic regression layer, ω k is the kth weight in the logistic regression layer, b k is the kth bias in the logistic regression layer, and σ is the Sigmoid function.
In the invention, a logistic regression layer processes a ripple characteristic vector to realize evaluation of one path of output.
The expression of the enhancement layer is as follows: Wherein h e,m is the stability after the m-th enhancement, h m is the stability of the m-th logistic regression layer output, L is the number of the logistic regression layers, and m is a positive integer.
The invention distributes the proportion coefficient to the output according to the output h m of each logistic regression layerThe output of each logistic regression layer is adjusted in a self-adaptive mode, and the attention of important features of the multi-input life prediction model is achieved in a self-adaptive mode.
The lifetime prediction layer has the expression: Wherein y is the service life of the power supply, h e,m is the stability after the m-th enhancement, omega m is the m-th weight in the logistic regression layer, b m is the m-th bias in the logistic regression layer, L is the number of the logistic regression layer, and m is a positive integer.
The life prediction layer disclosed by the invention is combined with a plurality of enhanced stabilities to comprehensively predict the life of a power supply.
The invention collects each path of voltage signal, screens out noise voltage value from each path of voltage signal, removes noise treatment, removes thermal noise and flicker noise, segments the noise-removed voltage signal according to peak positions to obtain multi-section voltage sub-signals, extracts a ripple characteristic vector from each section of voltage sub-signals, characterizes waveform shape characteristics of one section of voltage sub-signals, acquires ripple stability coefficient vectors according to the ripple characteristic vector of the same path, thus representing stability of ripple, reflects aging conditions of all parts of devices in a power supply through the stability of ripple, predicts service life based on a multi-input service life prediction model, realizes high-efficiency and real-time monitoring of running state of power supply equipment, and reduces missing detection or false alarm of faults.
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (1)
1. A multichannel combined power supply life monitoring and early warning system based on data analysis is characterized by comprising: the device comprises an acquisition unit, a denoising unit, a segmentation unit, a ripple characteristic extraction unit, a ripple stability coefficient extraction unit and a service life early warning unit;
the acquisition unit is used for acquiring voltage signals of each path of the multi-path combined power supply;
The denoising unit is used for screening noise voltage values from each path of voltage signal, and denoising to obtain denoising voltage signals;
The segmentation unit is used for segmenting the denoising voltage signal according to the peak position to obtain a plurality of segments of voltage sub-signals;
the ripple characteristic extraction unit is used for extracting a ripple characteristic vector for each section of voltage sub-signal;
The ripple stability coefficient extraction unit is used for obtaining a ripple stability coefficient vector according to the ripple characteristic vector of the same path;
the life early warning unit is used for predicting the life of the power supply based on a multi-input life prediction model according to the ripple stability coefficient vector, and early warning is carried out when the life of the power supply is lower than a threshold value;
the denoising unit includes: the device comprises a sliding queue, an average gradient calculation module, a noise voltage value screening module and a denoising module;
the sliding queue is N in length and is used for sliding on the voltage signal, filling voltage values in the voltage signal, and the sliding distance is 1 voltage value each time;
the average gradient calculation module is used for calculating an average gradient according to the voltage value in the sliding queue after each sliding of the sliding queue;
the noise voltage value screening module is used for classifying each voltage value into a noise voltage value when the gradient of the voltage value is larger than the average gradient;
the denoising module is used for denoising the noise voltage value to obtain a denoising voltage signal;
The formula for calculating the average gradient is as follows: Wherein d c is an average gradient, v i is the ith voltage value in the sliding queue, v i+1 is the (i+1) th voltage value in the sliding queue, i is a positive integer, and i is an absolute value;
the calculation formula of the gradient of the voltage value is as follows: Wherein d i is the gradient of the ith voltage value in the sliding queue, v i-1 is the ith-1 voltage value in the sliding queue;
The denoising processing formula of the denoising module is as follows: Wherein, the method comprises the steps of, wherein, For the denoised voltage value, v o is the noise voltage value, v l is the voltage value on the left side of the noise voltage value in the sliding queue, and v r is the voltage value on the right side of the noise voltage value in the sliding queue;
the ripple characteristic vector is: Wherein R j is the jth ripple feature vector, R j,1 is the 1 st element in the jth ripple feature vector, R j,2 is the 2 nd element in the jth ripple feature vector, R j,3 is the 3 rd element in the jth ripple feature vector, R j,4 is the 4 th element in the jth ripple feature vector, the 1 st element in the ripple feature vector is the peak-to-peak value, the 2 nd element in the ripple feature vector is the variance, the 3 rd element in the ripple feature vector is the skewness, the 4 th element in the ripple feature vector is the kurtosis, and j is a positive integer;
the ripple stability factor vector is: wherein S is a ripple stability coefficient vector, S 1 is the 1 st element in the ripple stability coefficient vector, S 2 is the 2 nd element in the ripple stability coefficient vector, S 3 is the 3 rd element in the ripple stability coefficient vector, S 4 is the 4 th element in the ripple stability coefficient vector, the 1 st element in the ripple stability coefficient vector is a peak-to-peak stability coefficient, the 2 nd element in the ripple stability coefficient vector is a variance stability coefficient, the 3 rd element in the ripple stability coefficient vector is a skewness stability coefficient, and the 4 th element in the ripple stability coefficient vector is a kurtosis stability coefficient;
the calculation formula of each element in the ripple stability coefficient vector is as follows: wherein s k is the kth element in the ripple stability coefficient vector, the value range of k is 1,2,3,4, r j,k is the kth element in the jth ripple characteristic vector, and M is the number of the ripple characteristic vectors in the same path;
The multiple-input life prediction model includes: a plurality of logistic regression layers, an enhancement layer, and a life prediction layer;
The input end of each logistic regression layer is used for inputting ripple stability coefficient vectors of each path, and the output end of each logistic regression layer is used for outputting stability;
The enhancement layer is used for enhancing the stability to obtain enhanced stability;
The life prediction layer is used for predicting the life of the power supply according to the stability after each enhancement;
the expression of each logistic regression layer is: Wherein h is the stability of the output of the logistic regression layer, omega k is the kth weight in the logistic regression layer, b k is the kth bias in the logistic regression layer, and sigma is the Sigmoid function;
the expression of the enhancement layer is as follows: Wherein h e,m is the stability of the m enhanced, h m is the stability of the m logistic regression layer output, L is the number of the logistic regression layers, and m is a positive integer;
The lifetime prediction layer has the expression: Wherein y is the service life of the power supply, h e,m is the stability after the m-th enhancement, omega m is the m-th weight in the logistic regression layer, b m is the m-th bias in the logistic regression layer, L is the number of the logistic regression layer, and m is a positive integer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410517918.7A CN118091478B (en) | 2024-04-28 | 2024-04-28 | Multichannel combined power supply life monitoring and early warning system based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410517918.7A CN118091478B (en) | 2024-04-28 | 2024-04-28 | Multichannel combined power supply life monitoring and early warning system based on data analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118091478A CN118091478A (en) | 2024-05-28 |
| CN118091478B true CN118091478B (en) | 2024-08-20 |
Family
ID=91163401
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410517918.7A Active CN118091478B (en) | 2024-04-28 | 2024-04-28 | Multichannel combined power supply life monitoring and early warning system based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118091478B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118275932B (en) * | 2024-06-04 | 2024-08-20 | 青岛元通电子有限公司 | Operation quality detection system suitable for ultralow noise power supply |
Family Cites Families (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002101668A (en) * | 2000-09-26 | 2002-04-05 | Meidensha Corp | Life time estimation method of semiconductor power converter and semiconductor power converter |
| JP2002107414A (en) * | 2000-09-29 | 2002-04-10 | Advantest Corp | Semiconductor testing device |
| WO2004099791A2 (en) * | 2003-04-23 | 2004-11-18 | Powertron Eng'g Co., Ltd | Diagnosis for expected life of emergency power apparatus |
| US10641837B2 (en) * | 2015-11-30 | 2020-05-05 | Linear Technology Corporation | Ripple monitoring |
| CN106528951B (en) * | 2016-10-18 | 2019-10-25 | 上海博强微电子有限公司 | A kind of method and apparatus of power battery life prediction and safe early warning |
| JP6665070B2 (en) * | 2016-10-25 | 2020-03-13 | 株式会社ミツバ | Motor control device and method for detecting current ripple of DC motor |
| CN110879254B (en) * | 2018-09-05 | 2021-09-17 | 哈尔滨工业大学 | Steel rail crack acoustic emission signal detection method based on improved least square generation type countermeasure network |
| CN110389290B (en) * | 2019-07-31 | 2022-05-06 | 东莞豪泽电子科技有限公司 | System, method and device for testing noise and evaluating service life of electronic component |
| CN113689044B (en) * | 2021-08-26 | 2024-06-21 | 北京航空航天大学 | Method and system for predicting residual service life of switching power supply |
| CN113705516B (en) * | 2021-09-03 | 2024-02-20 | 东南大学 | Sound source depth discrimination method based on radiation noise interference fringe characteristics |
| WO2023095171A1 (en) * | 2021-11-26 | 2023-06-01 | Bluesemi Research & Development Private Limited | Non-invasive sensor fusion and ai technology for human vital measurement |
| CN115728672A (en) * | 2022-04-02 | 2023-03-03 | 北京航空航天大学 | Circuit system health condition prediction method based on low-frequency noise and deep learning |
| CN116125327A (en) * | 2023-02-22 | 2023-05-16 | 湘潭大学 | Switching power supply fault diagnosis method based on improved wavelet denoising and LSTM |
| CN116520182B (en) * | 2023-05-11 | 2023-11-14 | 珠海中瑞电力科技有限公司 | DCS device power supply module abnormality early diagnosis method |
| CN116840720A (en) * | 2023-05-29 | 2023-10-03 | 武汉理工大学 | Fuel cell remaining life prediction method |
| CN116973761A (en) * | 2023-07-11 | 2023-10-31 | 黄河交通学院 | Method for predicting remaining service life of power battery based on particle swarm optimization |
| CN116626408B (en) * | 2023-07-25 | 2023-10-13 | 陕西威思曼高压电源股份有限公司 | Power supply ripple noise detection method based on machine learning |
| CN116973768A (en) * | 2023-09-13 | 2023-10-31 | 北京航空航天大学 | Lithium ion battery service life prediction method based on low-frequency electrochemical noise |
| CN117911792B (en) * | 2024-03-15 | 2024-06-04 | 垣矽技术(青岛)有限公司 | Pin detecting system for voltage reference source chip production |
-
2024
- 2024-04-28 CN CN202410517918.7A patent/CN118091478B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN118091478A (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN118091478B (en) | Multichannel combined power supply life monitoring and early warning system based on data analysis | |
| CN108008176B (en) | A kind of photovoltaic array real-time state monitoring and fault location system | |
| CN108229538B (en) | Vehicle system prediction apparatus and method | |
| CN105866638B (en) | A kind of city net cable connector state of insulation on-line monitoring prior-warning device and method | |
| CN108123849B (en) | Method, device, equipment and storage medium for determining threshold value for detecting network flow | |
| RU2484487C2 (en) | Method and device for determination of electric arc occurrence on at least one electric cable | |
| CN104221475B (en) | Fault detection, localization and performance monitoring of photosensors for lighting controls | |
| JP7157293B2 (en) | Failure detection/prediction device and failure detection/prediction program | |
| JPWO2009144820A1 (en) | Remote monitoring device | |
| JP2006300895A (en) | Facility monitoring method and facility monitoring device | |
| CN110084158B (en) | Electric equipment identification method based on intelligent algorithm | |
| CN116683648B (en) | Intelligent power distribution cabinet and control system thereof | |
| CN117078017A (en) | Intelligent decision analysis system for monitoring power grid equipment | |
| KR101631998B1 (en) | Solar power generation system having Controlling Device for Leakage Current and Double Protection for Surge | |
| Chai et al. | A non-intrusive deep learning based diagnosis system for elevators | |
| Paul et al. | Efficient-ArcNet: Series AC arc fault detection using lightweight convolutional neural network | |
| CN117767570A (en) | Lightning protection screen and intelligent monitoring system and method thereof | |
| CN109740797B (en) | Power equipment defect event early warning method based on conditional probability | |
| CN116714469A (en) | Charging pile health monitoring method, device, terminal and storage medium | |
| US9551744B2 (en) | Detecting early failures in printed wiring boards | |
| KR102429359B1 (en) | Main distrubution frame, mdf | |
| CN113514742B (en) | GIS partial discharge severity assessment method, system and medium | |
| CN110869607A (en) | Method, device and system for wind power converter management | |
| CN116961230B (en) | Intelligent power distribution cabinet for various occasions | |
| CN108226662B (en) | Airborne computer fault prediction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |