CN117129904B - Industrial power supply rapid switching monitoring method based on data analysis - Google Patents
Industrial power supply rapid switching monitoring method based on data analysis Download PDFInfo
- Publication number
- CN117129904B CN117129904B CN202311409965.1A CN202311409965A CN117129904B CN 117129904 B CN117129904 B CN 117129904B CN 202311409965 A CN202311409965 A CN 202311409965A CN 117129904 B CN117129904 B CN 117129904B
- Authority
- CN
- China
- Prior art keywords
- power supply
- data sequence
- voltage
- coefficient
- data
- 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 135
- 238000000034 method Methods 0.000 title claims abstract description 65
- 238000007405 data analysis Methods 0.000 title claims abstract description 12
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 15
- 230000001174 ascending effect Effects 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 9
- 230000035772 mutation Effects 0.000 claims description 7
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 4
- 238000009826 distribution Methods 0.000 abstract description 3
- 230000008859 change Effects 0.000 description 27
- 230000008569 process Effects 0.000 description 21
- 238000009776 industrial production Methods 0.000 description 13
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
Abstract
The invention relates to the technical field of data processing, and provides an industrial power supply rapid switching monitoring method based on data analysis, which comprises the following steps: acquiring industrial power supply state monitoring data; calculating a current overload state coefficient and a voltage overload state coefficient according to the industrial power supply state monitoring data; calculating a power supply overload state confidence coefficient according to the current overload state coefficient and the voltage overload state coefficient; acquiring a current overload trend coefficient and a voltage overload trend coefficient according to an empirical mode decomposition algorithm; calculating a power supply overload trend coefficient according to the power supply overload state confidence coefficient, the current overload trend coefficient and the voltage overload trend coefficient; and calculating a power supply switching weight coefficient according to the power supply overload trend coefficient, and acquiring an industrial power supply state monitoring data prediction result according to the power supply switching weight coefficient. According to the invention, the industrial power supply state monitoring data is subjected to weight distribution by calculating the power supply switching weight coefficient, so that the accuracy of predicting the industrial power supply state monitoring data is improved.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an industrial power supply rapid switching monitoring method based on data analysis.
Background
Along with the rapid development of industry, the process of industrial production gradually realizes an automatic production process, and the automatic industrial production improves the production efficiency, but a stable power supply system is required for maintaining the automatic industrial production, so that the stability requirement on a power system in the industrial production is also gradually improved. If the power supply system fails in the industrial production process, the operation of the whole industrial automatic production device can be stopped, and huge economic loss is brought to industrial production.
At present, an industrial power supply adopts a double-power supply system in industrial production, and the industrial power supply system is enabled to stably operate through switching between a main power supply and a standby power supply, so that stable operation of the industrial production is ensured. However, the power supply system adopting the dual power supplies needs to accurately judge the switching of the power supplies, and if the monitoring of the state of the power supplies deviates, the switching delay of the power supplies is caused, and the industrial production process is influenced. The power state is usually monitored by predicting the industrial power state monitoring data, so that economic loss caused by delay deviation of main power failure detection is avoided, but the prediction analysis of the power state monitoring data by the traditional moving weighted average prediction algorithm is affected by different data characteristic changes, so that the state prediction of the power supply is deviated, and the monitoring precision of quick switching of the industrial power supply is lower.
Disclosure of Invention
The invention provides a rapid switching monitoring method of an industrial power supply based on data analysis, which aims to solve the problem of low monitoring precision of rapid switching of the industrial power supply, and adopts the following technical scheme:
the invention provides a data analysis-based industrial power supply rapid switching monitoring method, which comprises the following steps of:
acquiring industrial power state monitoring data, wherein the monitoring data comprise a voltage time data sequence, a current time data sequence and a temperature time data sequence;
grouping the current time data sequence and the voltage time data sequence by adopting a mutation point detection algorithm; obtaining a maximum value point data sequence and a minimum value point data sequence of the current time data sequence according to the grouping result of the current time data sequence, and calculating a current overload state coefficient by utilizing the maximum value point data sequence and the minimum value point data sequence of the current time data sequence; obtaining a maximum value point data sequence and a minimum value point data sequence of the voltage time data sequence according to the grouping result of the voltage time data sequence, and calculating a voltage overload state coefficient by utilizing the maximum value point data sequence and the minimum value point data sequence of the voltage time data sequence; calculating a power supply overload state confidence coefficient of the industrial power supply according to the current overload state coefficient and the voltage overload state coefficient;
respectively acquiring a current overload trend coefficient and a voltage overload trend coefficient according to the voltage time data sequence and the current time data sequence; calculating the power supply overload trend coefficient of each group of industrial power supply state monitoring data according to the power supply overload state confidence coefficient, the current overload trend coefficient, the voltage overload trend coefficient and the temperature time data sequence; calculating a power supply switching weight coefficient of each group of industrial power supply state monitoring data according to the power supply overload trend coefficient of each group of industrial power supply state monitoring data; acquiring a predicted value of each group of industrial power state monitoring data according to a power switching weight coefficient of each group of industrial power state monitoring data;
calculating the state offset of the industrial power supply according to the predicted value of the state monitoring data of the industrial power supply; and monitoring the rapid switching of the industrial power supply according to the state offset of the industrial power supply.
Preferably, the method for grouping the current time data sequence and the voltage time data sequence by adopting the mutation point detection algorithm respectively comprises the following steps:
and respectively acquiring abrupt points in the current time data sequence and the voltage time data sequence by adopting an abrupt point detection algorithm, respectively taking the abrupt points in the current time data sequence and the voltage time data sequence as dividing points, and respectively grouping the current time data sequence and the voltage time data sequence according to the dividing points in the current time data sequence and the dividing points in the voltage time data sequence.
Preferably, the method for obtaining the maximum value point data sequence and the minimum value point data sequence of the current time data sequence according to the grouping result of the current time data sequence and calculating the current overload state coefficient by using the maximum value point data sequence and the minimum value point data sequence of the current time data sequence comprises the following steps:
taking each grouping result in the current time data sequence as a current data subsequence of the current data sequence, and taking a sequence formed by ordering data corresponding to all maximum value points in the current data subsequence according to a time ascending order as a maximum value point data sequence of the current data subsequence and the voltage data subsequence; acquiring absolute values of data corresponding to all minimum value points in the current data subsequence, and taking a sequence formed by sequencing all the absolute values according to a time ascending order as a minimum value point data sequence of the current data subsequence;
taking the difference value between the maximum value of an element in a maximum value point data sequence and the minimum value of an element in a minimum value point data sequence of any one current data subsequence in the current time data sequence as a first state coefficient of the any one current data subsequence; calculating the DTW distance of the maximum value point data sequence and the minimum value point data sequence of any one current data subsequence, and taking the calculation result of the DTW distance as a second state coefficient of any one current data subsequence; and taking the product of the first state coefficient and the second state coefficient of any one current data sub-sequence as the current overload state coefficient of any one current data sub-sequence.
Preferably, the method for obtaining the maximum value point data sequence and the minimum value point data sequence of the voltage time data sequence according to the grouping result of the voltage time data sequence and calculating the voltage overload state coefficient by using the maximum value point data sequence and the minimum value point data sequence of the voltage time data sequence comprises the following steps:
taking each grouping result in the voltage time data sequence as a voltage data subsequence of the voltage data sequence, and taking a sequence formed by ordering data corresponding to all maximum value points in the voltage data subsequence according to a time ascending order as the voltage data subsequence and the maximum value point data sequence of the voltage data subsequence; acquiring absolute values of data corresponding to all minimum value points in the voltage data subsequence, and taking a sequence formed by sequencing all the absolute values according to a time ascending order as a minimum value point data sequence of the voltage data subsequence;
taking the difference value between the maximum value of an element in a maximum value point data sequence and the minimum value of an element in a minimum value point data sequence of any one voltage data subsequence in the voltage time data sequence as a first state coefficient of the any one voltage data subsequence; calculating the DTW distance of the maximum value point data sequence and the minimum value point data sequence of any one voltage data subsequence, and taking the calculation result of the DTW distance as a second state coefficient of any one voltage data subsequence; and taking the product of the first state coefficient and the second state coefficient of any one of the voltage data sub-sequences as the voltage overload state coefficient of any one of the voltage data sub-sequences.
Preferably, the system according to the current overload stateThe method for calculating the power supply overload state confidence coefficient of the industrial power supply by the number and the voltage overload state coefficient comprises the following steps:
in the method, in the process of the invention,a power supply overload state confidence coefficient representing an industrial power supply; />A mean value of current overload state coefficients representing all current data sub-sequences in a current time data sequence, +.>Representing the average value of the voltage overload state coefficients of all the voltage data subsequences in the voltage time data sequence; />Representing the +.o in the current time data sequence>Current overload state coefficient of the individual current data subsequences, < ->Representing the +.>Voltage overload state coefficients for the individual voltage data subsequences; />Representing the number of current data sub-sequences in the current time data sequence,/->Representing the number of voltage data sub-sequences in the voltage time data sequence.
Preferably, the method for respectively obtaining the current overload trend coefficient and the voltage overload trend coefficient according to the voltage time data sequence and the current time data sequence comprises the following steps:
respectively acquiring residual components decomposed by the current time data sequence and residual components decomposed by the voltage time data sequence by adopting an empirical mode decomposition algorithm, calculating the difference value between the maximum value and the minimum value of the residual components decomposed by the current time data sequence, and taking the difference value as a current overload trend coefficient of the current time data sequence; and calculating the difference between the maximum value and the minimum value of the residual components decomposed by the voltage time data sequence, and taking the difference as the voltage overload trend coefficient of the voltage time data sequence.
Preferably, the method for calculating the power supply overload trend coefficient of each group of industrial power supply state monitoring data according to the power supply overload state confidence coefficient, the current overload trend coefficient, the voltage overload trend coefficient and the temperature time data sequence comprises the following steps:
in the method, in the process of the invention,indicate->Power supply overload trend coefficients of the group industrial power supply state monitoring data; />Indicate->A power overload state confidence coefficient for the group industrial power state monitoring data; />And->Respectively represent +.>Group and->Current overload trend coefficients of the group industrial power state monitoring data; />And->Respectively represent +.>Group and->Voltage overload trend coefficients of the group industrial power state monitoring data; />And->Respectively represent +.>Group and->The variation coefficient of the temperature time data sequence in the group industrial power state monitoring data; />Representing the number of groups of the acquired industrial power state monitoring data; />Indicating +.f. in all industrial power status monitoring data>Serial number of group industrial power status monitoring data.
Preferably, the method for calculating the power supply switching weight coefficient of each group of industrial power supply state monitoring data according to the power supply overload trend coefficient of each group of industrial power supply state monitoring data comprises the following steps:
and taking the power supply overload trend coefficient of any group of industrial power supply state monitoring data as a numerator, taking the sum of the power supply overload trend coefficients of all the industrial power supply state monitoring data as a denominator, and taking the ratio of the numerator to the denominator as the power supply switching weight coefficient of any group of industrial power supply state monitoring data.
Preferably, the method for obtaining the predicted value of each group of industrial power state monitoring data according to the power switching weight coefficient of each group of industrial power state monitoring data comprises the following steps:
and taking the power supply switching weight coefficient of each group of industrial power supply state monitoring data as the weight of each group of industrial power supply state monitoring data, and respectively acquiring a voltage predicted value, a current predicted value and a temperature predicted value of industrial power supply state monitoring by adopting a moving weighted average prediction algorithm.
Preferably, the method for calculating the industrial power state offset according to the predicted value of the industrial power state monitoring data comprises the following steps:
in the method, in the process of the invention,representing an industrial power state offset; />And->Respectively represent +.f in industrial power supply monitoring data>A predicted value and a threshold value of the seed monitoring data; />The type number of the collected industrial power supply state monitoring data is represented.
The beneficial effects of the invention are as follows: calculating a current overload state coefficient and a voltage overload state coefficient by analyzing local characteristics of current data and voltage data changes in a short time, and calculating a power supply overload state confidence coefficient according to the current overload state coefficient and the voltage overload state coefficient; and calculating a power supply overload trend coefficient according to the voltage and current change trend, the overall abnormal temperature degree and the power supply overload state confidence coefficient of the industrial power supply in a short time, and calculating a power supply switching weight coefficient according to the power supply overload trend coefficient. The method has the beneficial effects that local state characteristic differences in short time of the industrial power supply are considered, higher weight is given to the state monitoring data in the time period when the state of the industrial power supply is possibly changed, the accuracy of the prediction result of the state monitoring data of the industrial power supply is improved, and the accuracy of the rapid switching monitoring of the industrial power supply is further improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is a schematic flow chart of an industrial power supply rapid switching monitoring method based on data analysis according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a maximum value point data sequence and a minimum value point data sequence acquiring process according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of an industrial power supply rapid switching monitoring method based on data analysis according to an embodiment of the invention is shown, and the method includes the following steps:
step S001, acquiring industrial power supply state monitoring data.
In the industrial production process, a dual-power supply system is often adopted to ensure the stability of industrial operation, and when a main power supply fails, a standby power supply is switched to ensure the stability of the power supply system. Therefore, the switching of the industrial power supply is directly related to the stability of the power supply system, and the state of the industrial power supply needs to be accurately monitored, so that the quick switching response time of the industrial power supply is shorter, and the switching precision is higher. The method comprises the steps of acquiring voltage time data, current time data and temperature time data in the operation process of the industrial power supply by adopting a current sensor, a voltage sensor and a temperature sensor, and taking the voltage time data, the current time data and the temperature time data as industrial power supply state monitoring data.
The length of the data collection sequence of the three sensors is 600, and the time interval for collecting data is 1s, so that the collection time of each group of industrial power state monitoring data is 10min. Each group of industrial power state monitoring data comprises a voltage time data sequence, a current time data sequence and a temperature time data sequence, and the three data are collected at the same time at each collection time in the same group of industrial power state monitoring data collection process. Obtaining the working process of the industrial power supplyGroup continuous industrial power status monitoring data, +.>The size is 10, the practitioner can select proper data volume to analyze according to the industrial production precision requirement, and the median filtering algorithm is adopted to obtain +.>The group continuous industrial power state monitoring data is preprocessed, and the specific implementation process of the median filtering algorithm is a known technology and will not be described in detail.
To this end obtainTaking pretreatedA continuous set of industrial power status monitoring data.
Step S002, grouping is carried out according to the data characteristics of the industrial power supply state monitoring data, the current overload state coefficient and the voltage overload state coefficient are calculated according to grouping results, and the power supply overload state confidence coefficient is calculated according to the current overload state coefficient and the voltage overload state coefficient.
In the industrial production process, the power supply state may be affected by various interference factors, including the power supply self factor, the ambient temperature and other interference factors, which lead to the change of the industrial power supply state, the continuous operation may lead to the damage and failure of the industrial power supply, meanwhile, the inaccurate detection of the power supply failure may also lead to the longer switching response time of the industrial power supply, so that the stability of the power supply system is reduced, and the short power supply stop may be caused when serious, which affects the industrial production progress.
Thus, industrial power state monitoring data are collected: the current time data, the voltage time data and the temperature time data are composed of a current time data sequence, a voltage time data sequence and a temperature time data sequence according to the group of industrial power state monitoring data in the step S001. The state of the industrial power supply may be characterized by overload current, overload voltage or abnormal fluctuation of voltage and current, so that the local characteristic of the data is analyzed in the acquisition time of each group of industrial power supply state monitoring data, and the state characteristic of the industrial power supply in the time period is reflected by the local characteristic.
Specifically, a current time data sequence and a voltage time data sequence are input, mutation data of the current time data sequence and mutation data of the voltage time data sequence are respectively obtained by processing by adopting a Bernaola Galvan segmentation algorithm, the mutation data in the two sequences are taken as segmentation points to respectively group the two sequences, each group of data in the current time data sequence and the voltage time data sequence is respectively taken as a current data subsequence of the current time data sequence and a voltage data subsequence of the voltage time data sequence, and Bernaola is used as a voltage data subsequence of the current time data sequenceThe Galvan segmentation algorithm is a well-known technique, and the specific calculation process will not be described in detail. Based on the groupings of the current time data series and the voltage time data series, local characteristics of the state of the industrial power supply in a short time are analyzed. With the 1 st current data sub-sequence in the current time data sequenceFor example, will->The data corresponding to the maximum point in (a) are ordered in a time-ascending order to form a sequence as +.>Maximum value point data sequence of +.>The method comprises the steps of carrying out a first treatment on the surface of the Separately obtain->Absolute value of data corresponding to each minimum point of (a) is to be +.>All of the absolute values of (a) are ordered in ascending order of time as +.>Is>The specific maximum value point and minimum value point acquisition process is shown in fig. 2.
Since the state of the industrial power supply changes when the state of the industrial power supply changes, the current output by the industrial power supply is generally stable alternating currentAnd->Is the root of the change of the differential reaction current change stateThe current overload state coefficient is calculated according to the change of the current state, and the specific calculation process is as follows:
in the method, in the process of the invention,representing the +.o in the current time data sequence>Current overload state coefficients for the respective current data subsequences;and->Respectively representing the +.sup.th of the current time data sequence>A maximum point data sequence and a minimum point data sequence of the individual current data subsequences; />And->Respectively indicate->Maximum value sum->Is the minimum value of (a);representing the calculation->And->DTW distance of (2); wherein->For the first state coefficient->Is the second state coefficient.
If the state of the output current change of the industrial power supply is changed and the state change is more complex, the calculated result is thatThe larger the value of (2), the more obvious overload trend appears in the current change, and the calculated value isThe larger the value of (2), the overload state coefficient of the resulting current time data sequence +.>The larger the value of (c) is, the greater the possibility that the output current variation of the industrial power supply is overloaded and unstable.
Further, the output current and output voltage of the power supply can be influenced by the change of the state of the industrial power supply, and meanwhile, the abnormal operation of the power supply is caused by the change of the state of the industrial power supply, and the phenomenon of high temperature of the power supply is also possible to occur. The status characteristics of the industrial power supply can be analyzed by overload conditions and power supply temperatures of the industrial power supply status monitoring data for different time periods. Firstly, calculating a power supply overload state confidence coefficient according to the change characteristics of the output voltage and the output current of the industrial power supply.
Specifically, the same calculation mode as the current overload state coefficient of the subsequence in the current time data sequence is adopted, a maximum value point data sequence and a minimum value point data sequence of each voltage data subsequence in the voltage data sequence are respectively obtained, and the difference value of the maximum value of the element in the maximum value point data sequence and the minimum value of the element in the minimum value point data sequence of each voltage data subsequence in the voltage time data sequence is used as the first state coefficient of each voltage data subsequence; calculating the DTW distance of the maximum value point data sequence and the minimum value point data sequence of each voltage data subsequence, and taking the calculation result of the DTW distance as a second state coefficient of each voltage data subsequence; taking the product of the first state coefficient and the second state coefficient of each voltage data sub-sequence as the voltage overload state coefficient of each voltage data sub-sequence.
Calculating a power supply overload state confidence coefficient according to the difference between the current overload state coefficient of the current data subsequence in the current time data sequence and the voltage overload state coefficient of the voltage data subsequence in the voltage time data sequence, wherein the specific calculation formula is as follows:
in the method, in the process of the invention,a power supply overload state confidence coefficient representing an industrial power supply; />A mean value of current overload state coefficients representing all current data sub-sequences in a current time data sequence, +.>Representing the average value of the voltage overload state coefficients of all the voltage data subsequences in the voltage time data sequence; />Representing the +.o in the current time data sequence>Current overload state coefficient of the individual current data subsequences, < ->Representing the +.>Voltage overload state coefficients for the individual voltage data subsequences; />Representing the number of current data sub-sequences in the current time data sequence,/->Representing the number of voltage data sub-sequences in the voltage time data sequence.
If the state change of the voltage and the current of the industrial power supply is large, the calculated state is obtainedIs larger, and the voltage and current variation of the industrial power supply are larger, calculated +.>The larger the value of (2), the resulting power supply overload state confidence coefficient +.>The greater the value of (c) is, the greater the likelihood of an overload condition occurring in the industrial power supply from a local characterization of the output voltage and output current of the industrial power supply.
Thus, the power supply overload state confidence coefficient of the industrial power supply is obtained.
Step S003, a current overload trend coefficient and a voltage overload trend coefficient are obtained according to an empirical mode decomposition algorithm, a power overload trend coefficient is calculated according to a power overload state confidence coefficient, the current overload trend coefficient and the voltage overload trend coefficient, a power switching weight coefficient is calculated according to the power overload trend coefficient, and an industrial power state monitoring data prediction result is obtained according to the power switching weight coefficient.
In step S002, the change of the industrial power state is analyzed by collecting the local features of the industrial power state monitoring data each time, further, the overall trend of the industrial power state change is analyzed according to the global features of the industrial power state monitoring data, and the power overload trend coefficient of the industrial power is calculated by the analysis results of the local features and the global features.
Specifically, the input is a current time data sequence and a voltage time data sequence, residual components of the current time data sequence and the voltage time data sequence are respectively obtained by adopting an empirical mode decomposition algorithm, and a specific calculation process of the empirical mode decomposition is a known technology and will not be described in detail. Residual components obtained by empirical mode decomposition are components of which input data can not be decomposed any more and finally show monotonicity, the integral change characteristics of the two data can be analyzed through the residual components of the current time data sequence and the voltage time data sequence, absolute values of differences between maximum values and minimum values of the residual components of the current time data sequence and the voltage time data sequence are calculated respectively, and the absolute values are used as current overload trend coefficients of the current time data sequence and voltage overload trend coefficients of the voltage time data sequence respectively.
Further, the abnormal operation of the power supply caused by the change of the state of the industrial power supply may occur due to the higher temperature of the power supply, and the variation coefficient of the temperature time data sequence is calculated according to the overall characteristic of the temperature change of the analysis of the temperature characteristic of the state change of the industrial power supplyThe specific calculation process of the variation coefficient is a known technology and will not be described in detail. Calculating a power supply overload trend coefficient according to the power supply overload state confidence coefficient of the industrial power supply state change and the global characteristic of the industrial power supply state monitoring data, wherein the specific calculation formula is as follows:
in the method, in the process of the invention,indicate->Power supply overload trend coefficients of the group industrial power supply state monitoring data; />Indicate->A power overload state confidence coefficient for the group industrial power state monitoring data; />And->Respectively represent +.>Group and->Current overload trend coefficients of the group industrial power state monitoring data; />And->Respectively represent +.>Group and->Voltage overload trend coefficients of the group industrial power state monitoring data; />And->Respectively represent +.>Group and->The variation coefficient of the temperature time data sequence in the group industrial power state monitoring data; />Representing the number of groups of the acquired industrial power state monitoring data;/>indicating +.f. in all industrial power status monitoring data>Serial number of group industrial power status monitoring data.
If the industrial power supply is at the firstThe overall change trend in the group industrial power supply state monitoring data acquisition time period is greatly different from the overall change trend in other time periods, and the overall change trend is calculatedThe larger the value of (2), and the state of the industrial power supply is +.>The greater the likelihood of a state change during a group industrial power state monitoring data acquisition period, the +.>The larger the value of the power supply overload trend coefficient finally calculated, the larger the value of the power supply overload trend coefficient is, which is shown in the +.>The greater the likelihood that an industrial power state changes and an overload condition occurs within a time period of collection of the set of industrial power state monitoring data, the greater the correlation to the predictions of the industrial power state monitoring data.
Input being takenProcessing the industrial power state monitoring data by adopting a moving weighted average prediction algorithm to obtain a voltage predicted value, a current predicted value and a temperature predicted value of industrial power state monitoring, wherein the distribution of the weight of each group of industrial power state monitoring data is calculated according to the power overload trend coefficient, and the specific calculation is performedThe process is as follows: />
In the method, in the process of the invention,indicate->Power supply switching weight coefficients for the group industrial power supply status monitoring data; />And->Respectively represent +.>Group and->Power supply overload trend coefficients for group industrial power supply status monitoring data. If%>The greater the possibility that the industrial power state changes and an overload condition occurs within the period of time of the collection of the industrial power state monitoring data, the more the industrial power state is monitoredThe group industrial power status monitoring data is given a greater weight. The input is industrial power state monitoring data, and according to the distribution of the weights of each group of industrial power state monitoring data, a moving weighted average prediction algorithm is adopted to obtain a voltage predicted value of industrial power state monitoring +.>Current prediction value->And temperature prediction value->。
Thus, the voltage predicted value, the current predicted value and the temperature predicted value of the industrial power supply state monitoring are obtained.
And S004, analyzing the state of the industrial power supply according to the prediction result of the industrial power supply state monitoring data, and completing the rapid switching detection of the industrial power supply according to the analysis result.
Obtaining a voltage predicted value, a current predicted value and a temperature predicted value of the industrial power state monitoring according to the calculation in the step S003, wherein the voltage predicted value, the current predicted value and the temperature predicted value are respectively、/>、/>The method comprises the steps of carrying out a first treatment on the surface of the Acquiring a voltage threshold value, a current threshold value and a temperature threshold value according to the overload protection related regulation of the industrial power supply, wherein the voltage threshold value, the current threshold value and the temperature threshold value are respectively +.>、/>、/>. Calculating the state offset of the industrial power supply according to the obtained predicted value, wherein the specific calculation process is as follows: />
In the method, in the process of the invention,representing an industrial power state offset; />And->Respectively represent industrial electricityFirst->A predicted value and a threshold value of the seed monitoring data; />The number of types of the collected industrial power supply state monitoring data is represented, and the size of the collected industrial power supply state monitoring data takes an empirical value of 3.
Further, setting the industrial power supply state change threshold to be 0.1, and calculating the calculated industrial power supply state offsetAnd comparing the industrial power supply state change threshold value with the industrial power supply state change threshold value, and if the industrial power supply state offset is larger than the industrial power supply state change threshold value, sending a switching signal to the industrial power supply rapid switching device to finish rapid switching of the industrial power supply.
Thus, the rapid switching monitoring of the industrial power supply is completed.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (4)
1. The industrial power supply rapid switching monitoring method based on data analysis is characterized by comprising the following steps of:
acquiring industrial power state monitoring data, wherein the monitoring data comprise a voltage time data sequence, a current time data sequence and a temperature time data sequence;
grouping the current time data sequence and the voltage time data sequence by adopting a mutation point detection algorithm; obtaining a maximum value point data sequence and a minimum value point data sequence of the current time data sequence according to the grouping result of the current time data sequence, and calculating a current overload state coefficient by utilizing the maximum value point data sequence and the minimum value point data sequence of the current time data sequence; obtaining a maximum value point data sequence and a minimum value point data sequence of the voltage time data sequence according to the grouping result of the voltage time data sequence, and calculating a voltage overload state coefficient by utilizing the maximum value point data sequence and the minimum value point data sequence of the voltage time data sequence; calculating a power supply overload state confidence coefficient of the industrial power supply according to the current overload state coefficient and the voltage overload state coefficient;
respectively acquiring a current overload trend coefficient and a voltage overload trend coefficient according to the voltage time data sequence and the current time data sequence; calculating the power supply overload trend coefficient of each group of industrial power supply state monitoring data according to the power supply overload state confidence coefficient, the current overload trend coefficient, the voltage overload trend coefficient and the temperature time data sequence; calculating a power supply switching weight coefficient of each group of industrial power supply state monitoring data according to the power supply overload trend coefficient of each group of industrial power supply state monitoring data; acquiring a predicted value of each group of industrial power state monitoring data according to a power switching weight coefficient of each group of industrial power state monitoring data;
calculating the state offset of the industrial power supply according to the predicted value of the state monitoring data of the industrial power supply; monitoring the rapid switching of the industrial power supply according to the state offset of the industrial power supply;
the method for obtaining the maximum value point data sequence and the minimum value point data sequence of the current time data sequence according to the grouping result of the current time data sequence and calculating the current overload state coefficient by utilizing the maximum value point data sequence and the minimum value point data sequence of the current time data sequence comprises the following steps:
taking each grouping result in the current time data sequence as a current data subsequence of the current data sequence, and taking a sequence formed by ordering data corresponding to all maximum value points in the current data subsequence according to a time ascending order as a maximum value point data sequence of the current data subsequence and the voltage data subsequence; acquiring absolute values of data corresponding to all minimum value points in the current data subsequence, and taking a sequence formed by sequencing all the absolute values according to a time ascending order as a minimum value point data sequence of the current data subsequence;
taking the difference value between the maximum value of an element in a maximum value point data sequence and the minimum value of an element in a minimum value point data sequence of any one current data subsequence in the current time data sequence as a first state coefficient of the any one current data subsequence; calculating the DTW distance of the maximum value point data sequence and the minimum value point data sequence of any one current data subsequence, and taking the calculation result of the DTW distance as a second state coefficient of any one current data subsequence; taking the product of the first state coefficient and the second state coefficient of any one current data sub-sequence as the current overload state coefficient of any one current data sub-sequence;
the method for obtaining the maximum value point data sequence and the minimum value point data sequence of the voltage time data sequence according to the grouping result of the voltage time data sequence and calculating the voltage overload state coefficient by utilizing the maximum value point data sequence and the minimum value point data sequence of the voltage time data sequence comprises the following steps:
taking each grouping result in the voltage time data sequence as a voltage data subsequence of the voltage data sequence, and taking a sequence formed by ordering data corresponding to all maximum value points in the voltage data subsequence according to a time ascending order as the voltage data subsequence and the maximum value point data sequence of the voltage data subsequence; acquiring absolute values of data corresponding to all minimum value points in the voltage data subsequence, and taking a sequence formed by sequencing all the absolute values according to a time ascending order as a minimum value point data sequence of the voltage data subsequence;
taking the difference value between the maximum value of an element in a maximum value point data sequence and the minimum value of an element in a minimum value point data sequence of any one voltage data subsequence in the voltage time data sequence as a first state coefficient of the any one voltage data subsequence; calculating the DTW distance of the maximum value point data sequence and the minimum value point data sequence of any one voltage data subsequence, and taking the calculation result of the DTW distance as a second state coefficient of any one voltage data subsequence; taking the product of the first state coefficient and the second state coefficient of any one of the voltage data sub-sequences as the voltage overload state coefficient of any one of the voltage data sub-sequences;
the method for calculating the power supply overload state confidence coefficient of the industrial power supply according to the current overload state coefficient and the voltage overload state coefficient comprises the following steps:in (1) the->A power supply overload state confidence coefficient representing an industrial power supply; />Representing the average of the current overload state coefficients of all current data sub-sequences in the current time data sequence,representing the average value of the voltage overload state coefficients of all the voltage data subsequences in the voltage time data sequence; />Representing the +.o in the current time data sequence>Current overload state coefficient of the individual current data subsequences, < ->Representing the +.>Voltage overload state coefficients for the individual voltage data subsequences; />Representing current data subsequences in a current time data sequenceNumber of columns>Representing the number of voltage data sub-sequences in the voltage time data sequence;
the method for respectively obtaining the current overload trend coefficient and the voltage overload trend coefficient according to the voltage time data sequence and the current time data sequence comprises the following steps:
respectively acquiring residual components decomposed by the current time data sequence and residual components decomposed by the voltage time data sequence by adopting an empirical mode decomposition algorithm, calculating the difference value between the maximum value and the minimum value of the residual components decomposed by the current time data sequence, and taking the difference value as a current overload trend coefficient of the current time data sequence; calculating a difference value between a maximum value and a minimum value of residual components decomposed by the voltage time data sequence, and taking the difference value as a voltage overload trend coefficient of the voltage time data sequence;
the method for calculating the power supply overload trend coefficient of each group of industrial power supply state monitoring data according to the power supply overload state confidence coefficient, the current overload trend coefficient, the voltage overload trend coefficient and the temperature time data sequence comprises the following steps:in (1) the->Indicate->Power supply overload trend coefficients of the group industrial power supply state monitoring data; />Indicate->A power overload state confidence coefficient for the group industrial power state monitoring data; />And->Respectively represent +.>Group and->Current overload trend coefficients of the group industrial power state monitoring data; />And->Respectively represent +.>Group and->Voltage overload trend coefficients of the group industrial power state monitoring data;and->Respectively represent +.>Group and->The variation coefficient of the temperature time data sequence in the group industrial power state monitoring data; />Representing the number of groups of the acquired industrial power state monitoring data; />Indicating +.f. in all industrial power status monitoring data>Serial number of group industrial power status monitoring data;
the method for calculating the power supply switching weight coefficient of each group of industrial power supply state monitoring data according to the power supply overload trend coefficient of each group of industrial power supply state monitoring data comprises the following steps:
and taking the power supply overload trend coefficient of any group of industrial power supply state monitoring data as a numerator, taking the sum of the power supply overload trend coefficients of all the industrial power supply state monitoring data as a denominator, and taking the ratio of the numerator to the denominator as the power supply switching weight coefficient of any group of industrial power supply state monitoring data.
2. The method for monitoring the rapid switching of the industrial power supply based on the data analysis according to claim 1, wherein the method for grouping the current time data sequence and the voltage time data sequence by adopting the mutation point detection algorithm is characterized in that:
and respectively acquiring abrupt points in the current time data sequence and the voltage time data sequence by adopting an abrupt point detection algorithm, respectively taking the abrupt points in the current time data sequence and the voltage time data sequence as dividing points, and respectively grouping the current time data sequence and the voltage time data sequence according to the dividing points in the current time data sequence and the dividing points in the voltage time data sequence.
3. The method for rapidly switching and monitoring industrial power supply based on data analysis according to claim 1, wherein the method for obtaining the predicted value of each group of industrial power supply state monitoring data according to the power supply switching weight coefficient of each group of industrial power supply state monitoring data comprises the following steps:
and taking the power supply switching weight coefficient of each group of industrial power supply state monitoring data as the weight of each group of industrial power supply state monitoring data, and respectively acquiring a voltage predicted value, a current predicted value and a temperature predicted value of industrial power supply state monitoring by adopting a moving weighted average prediction algorithm.
4. The method for monitoring the rapid switching of the industrial power supply based on the data analysis according to claim 1, wherein the method for calculating the state offset of the industrial power supply according to the predicted value of the state monitoring data of the industrial power supply comprises the following steps:in (1) the->Representing an industrial power state offset; />And->Respectively represent +.f in industrial power supply monitoring data>A predicted value and a threshold value of the seed monitoring data; />The type number of the collected industrial power supply state monitoring data is represented.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311409965.1A CN117129904B (en) | 2023-10-27 | 2023-10-27 | Industrial power supply rapid switching monitoring method based on data analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311409965.1A CN117129904B (en) | 2023-10-27 | 2023-10-27 | Industrial power supply rapid switching monitoring method based on data analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117129904A CN117129904A (en) | 2023-11-28 |
CN117129904B true CN117129904B (en) | 2023-12-22 |
Family
ID=88854984
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311409965.1A Active CN117129904B (en) | 2023-10-27 | 2023-10-27 | Industrial power supply rapid switching monitoring method based on data analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117129904B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB9407933D0 (en) * | 1994-01-28 | 1994-06-15 | Kb Electronics 1989 Ltd | Bimodal fast transfer off-line uninterruptible power supply |
RU2133542C1 (en) * | 1998-02-18 | 1999-07-20 | Казьмин Григорий Павлович | Method controlling system of uninterrupted power supply under emergency conditions |
CN101674019A (en) * | 2008-03-24 | 2010-03-17 | 技领半导体(上海)有限公司 | Programmable integrated circuit and method |
CN211790945U (en) * | 2020-04-28 | 2020-10-27 | 浙江泰达尔智能科技有限公司 | Novel low-voltage direct-current standby power supply switching module |
CN113904428A (en) * | 2021-09-03 | 2022-01-07 | 杭州电子科技大学 | Uninterrupted power supply system and method in power protection area |
CN115754790A (en) * | 2022-11-16 | 2023-03-07 | 国网天津市电力公司电力科学研究院 | Power supply side fault diagnosis and automatic switching method for low-voltage dual-power system |
-
2023
- 2023-10-27 CN CN202311409965.1A patent/CN117129904B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB9407933D0 (en) * | 1994-01-28 | 1994-06-15 | Kb Electronics 1989 Ltd | Bimodal fast transfer off-line uninterruptible power supply |
RU2133542C1 (en) * | 1998-02-18 | 1999-07-20 | Казьмин Григорий Павлович | Method controlling system of uninterrupted power supply under emergency conditions |
CN101674019A (en) * | 2008-03-24 | 2010-03-17 | 技领半导体(上海)有限公司 | Programmable integrated circuit and method |
CN211790945U (en) * | 2020-04-28 | 2020-10-27 | 浙江泰达尔智能科技有限公司 | Novel low-voltage direct-current standby power supply switching module |
CN113904428A (en) * | 2021-09-03 | 2022-01-07 | 杭州电子科技大学 | Uninterrupted power supply system and method in power protection area |
CN115754790A (en) * | 2022-11-16 | 2023-03-07 | 国网天津市电力公司电力科学研究院 | Power supply side fault diagnosis and automatic switching method for low-voltage dual-power system |
Also Published As
Publication number | Publication date |
---|---|
CN117129904A (en) | 2023-11-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109814527B (en) | Industrial equipment fault prediction method and device based on LSTM recurrent neural network | |
CN112946484B (en) | SOC estimation method, system, terminal equipment and readable storage medium based on BP neural network | |
TW201915727A (en) | Fault detection and classification method of multi-sensors | |
CN109389314B (en) | Quality soft measurement and monitoring method based on optimal neighbor component analysis | |
CN108074015B (en) | Ultra-short-term prediction method and system for wind power | |
JPWO2015136586A1 (en) | Factor analysis apparatus, factor analysis method, and factor analysis program | |
CN115902647B (en) | Intelligent battery state monitoring method | |
CN111368428B (en) | Sensor precision degradation fault detection method based on monitoring second-order statistics | |
CN111177970B (en) | Multi-stage semiconductor process virtual metering method based on Gaussian process and convolutional neural network | |
CN113420061A (en) | Steady state working condition analysis method, optimization debugging method and system of oil refining and chemical production device | |
CN117235462B (en) | Intelligent fault prediction method for bag type packaging machine based on time sequence data analysis | |
CN116258281A (en) | Internet of things fire control monitoring and regulating system based on cloud platform management | |
CN116955895B (en) | Transformer operation abnormity monitoring method and system | |
CN117390591B (en) | Operation monitoring method and system for coal conveying belt sampling machine based on electric parameter analysis | |
CN117129904B (en) | Industrial power supply rapid switching monitoring method based on data analysis | |
CN116772285B (en) | Intelligent building heating load safety real-time monitoring method | |
CN117289168A (en) | Product quality monitoring system based on sensor data and intelligent algorithm | |
CN107633271B (en) | Method for calculating inaccurate probability of steady-state availability of power system | |
CN111913463B (en) | State monitoring method for chemical volume control system of nuclear power plant | |
CN113255207A (en) | Iterative multi-output-Markov chain-based multi-step prediction method for argon fraction variable of air separation system | |
CN117150244B (en) | Intelligent power distribution cabinet state monitoring method and system based on electrical parameter analysis | |
CN115877890B (en) | Temperature control method and system for CVD equipment | |
CN110322063A (en) | A kind of power consumption simulated prediction method and storage medium | |
CN117313014B (en) | Real-time monitoring and early warning method for abnormal energy consumption data of kiln operation | |
CN116757337B (en) | House construction progress prediction system based on artificial intelligence |
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 |