CN117783745B - Data online monitoring method and system for battery replacement cabinet - Google Patents
Data online monitoring method and system for battery replacement cabinet Download PDFInfo
- Publication number
- CN117783745B CN117783745B CN202311836916.6A CN202311836916A CN117783745B CN 117783745 B CN117783745 B CN 117783745B CN 202311836916 A CN202311836916 A CN 202311836916A CN 117783745 B CN117783745 B CN 117783745B
- Authority
- CN
- China
- Prior art keywords
- data
- data point
- data points
- reference data
- obtaining
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000012544 monitoring process Methods 0.000 title claims abstract description 34
- 238000012937 correction Methods 0.000 claims abstract description 42
- 230000008859 change Effects 0.000 claims abstract description 25
- 230000002159 abnormal effect Effects 0.000 claims abstract description 16
- 238000010586 diagram Methods 0.000 claims description 22
- 238000005192 partition Methods 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000006243 chemical reaction Methods 0.000 claims description 17
- 238000009826 distribution Methods 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 6
- 230000003247 decreasing effect Effects 0.000 description 5
- 238000005457 optimization Methods 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Landscapes
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
- Tests Of Electric Status Of Batteries (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to a data online monitoring method and system for a battery-changing cabinet, comprising the following steps: collecting current data of a battery-changing cabinet, and selecting an initial window length of a reference data point; obtaining the preference degree of an initial window of the reference data points; obtaining a first correction factor and a second correction factor of the preference degree; obtaining final preference degree of the initial window according to the preference degree, the first correction factor and the second correction factor, and obtaining the optimal window length according to the final preference degree; constructing a box line graph according to the optimal window length to obtain abnormal data points, and obtaining the noise influence degree of the reference data points according to the number of the abnormal data points and the change of the local range data current value; removing noise points according to a preset threshold value; therefore, when the working state of the battery-changing cabinet is monitored according to the denoised data, the change of the current data can be accurately reflected, and the monitoring result is more accurate.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a data online monitoring method and system for a battery exchange cabinet.
Background
With the popularization of electric vehicles, the battery changing cabinet system becomes more and more important, so that the problem of limitation of the endurance mileage of the electric vehicles can be solved, the use convenience of the electric vehicles is improved, and the requirement of a user on battery maintenance is reduced. The battery changing cabinet is a device for battery charging and changing service of electric vehicles such as electric bicycles, electric automobiles and the like. In a battery changing cabinet system, a user may install a battery on a vehicle and then use an automated system in the device to charge or change the battery to extend the range of the vehicle. In order to ensure the reliability, efficiency and safety of the power conversion cabinet system, online data monitoring is required. The monitoring data includes information such as state of charge, battery temperature, battery capacity, device status, user interaction, etc. The online data monitoring can help operators to find problems in time and take measures to improve the operation efficiency and the safety of the system. However, when monitoring data, because noise exists in the collected current data, denoising processing needs to be performed on the collected data.
In the prior art, a common denoising method is to detect noise points through a box diagram and then perform interpolation operation. However, in the process of constructing the box diagram, the fixed box diagram length (the data set participating in constructing the box diagram) may have different effects on different data distributions, that is, the denoising effect is inconsistent, and the denoising effect is good or bad, so that the denoising effect is affected. Because the acquired power conversion cabinet is time sequence data, the time sequence data has different change trends at different moments, namely the number of the data meeting the unified box diagram at different moments is different, so that the length of the self-adaptive box diagram is required.
Disclosure of Invention
In order to solve the problems, the invention provides a data on-line monitoring method and system for a battery-changing cabinet.
The invention discloses a data on-line monitoring method for a battery-changing cabinet, which adopts the following technical scheme:
The embodiment of the invention provides a data online monitoring method for a battery changing cabinet, which comprises the following steps of:
collecting current data of a battery changing cabinet; the power conversion cabinet comprises a plurality of power conversion cabinet bins, the power conversion cabinet current data comprise a plurality of data points, and the data points represent current values collected by each power conversion cabinet bin at each moment;
Marking any one data point as a reference data point, and selecting the initial window length of the reference data point; obtaining a plurality of bit intervals according to the data changes contained in the initial window; obtaining the preference degree of an initial window of the reference data point according to the first-order difference and the second-order difference of the data in the bit interval; obtaining a first correction factor of the preference degree according to the time sequence distribution and the data change of the bit intervals; obtaining a second correction factor of the optimal degree according to the difference value of the current value and the total current value of each battery changing cabinet bin; obtaining final preference degree of the initial window of the reference data point according to the preference degree, the first correction factor and the second correction factor, and obtaining the optimal window length of the reference data point according to the final preference degree;
Constructing a box diagram according to the optimal window length to obtain abnormal data points, obtaining local range data of reference data points, and obtaining noise influence degree of the reference data points according to the variation of current values of the abnormal data points and the local range data; removing noise points according to the magnitude relation between the noise influence degree and a preset threshold value; and monitoring the working state of the battery exchange cabinet according to the denoised data.
Further, the step of recording any one data point as a reference data point and selecting the initial window length of the reference data point comprises the following specific steps:
and constructing an initial window with the length of a by taking the reference data point as the center, wherein the initial window contains a data points, and a is a preset integer value.
Further, the obtaining a plurality of bit intervals according to the data changes contained in the initial window includes the following specific steps:
The data points in the initial window are sequenced from small to large and recorded as a first data sequence, and the first data sequence is equally divided into eight equal divisions to obtain eight bit intervals; the interval from the minimum value to the first octant is recorded as a first quantile interval, the interval from the first octant to the second octant is recorded as a second quantile interval, and the like, so that eight quantile intervals are obtained.
Further, the obtaining the preference degree of the initial window of the reference data point according to the first-order difference and the second-order difference of the data in the bit interval includes the following specific steps:
respectively obtaining the average value of the current values of all data points in eight bit intervals, arranging the average value of the current values in each bit interval from large to small to obtain a second sequence, and then obtaining the median of the second sequence;
Acquiring a first-order difference and a second-order difference of data points in each bit interval, wherein the first-order difference is obtained by subtracting the average value of the current value of each data point in the first bit interval from the average value of the current value of each data point in the second bit interval, and the average value of the current value of each data point in the first bit interval represents the first-order difference; subtracting the average value of the current value of each data point in the second bit interval from the average value of the current value of each data point in the third bit interval to represent a second first step difference; the second step is a second first step minus the first step, representing the first second step; subtracting the second first step from the third first step represents a second step; the calculation formula for the preference of the initial window of reference data points is as follows:
Where gamma t denotes the preference degree of the t-th reference data point corresponding to the initial window, gamma 1 denotes the preference degree of the bit section included on the side smaller than the median, gamma 2 denotes the preference degree of the bit section included on the side larger than the median, Representing a first level difference of the ith partition zone smaller than the median side,/>Representing the second order difference of the ith partition zone smaller than the median side,/>Represents the average interval of adjacent current data points in the ith partition less than the median side corresponding to time,/>Representing a first level difference of a j-th partition greater than the median side,/>Representing the second order difference of the j-th partition greater than the median side,/>Then the average interval of the corresponding time of the adjacent current data points in the j-th bit interval on the side greater than the median is represented, y represents the number of bit intervals contained on the side less than the median, exp () represents an exponential function based on a natural constant.
Further, the obtaining the first correction factor of the preference degree according to the time sequence distribution and the data change of the bit interval comprises the following specific steps:
Where, δ 1 represents a first correction factor of the preference of the initial window of reference data points, Representing the average interval of adjacent current data points in the ith partition corresponding to time,/>Representing the average interval of the time intervals corresponding to the two largest adjacent current data points removed from the ith partition.
Further, the obtaining the second correction factor of the preference degree according to the difference value between the current value and the total current value of each battery changing cabinet bin comprises the following specific steps:
Wherein delta 2 represents a second correction factor of the preference degree of the initial window of the reference data point, I i represents the total current of the moment corresponding to the ith data point in the window, I i1 represents the current value of the 1 st battery changing cabinet bin running at the moment corresponding to the ith data point,
I i2 represents the current value of the 2 nd power change cabinet bin which is operated at the moment corresponding to the ith data point, I im represents the current value of the m th power change cabinet bin which is operated at the moment corresponding to the ith data point, N represents the window length, the total current represents the sum of the current values of all the power change cabinet bins at the corresponding moment, and m represents the number of the power change cabinet bins.
Further, the obtaining the final preference degree of the initial window of the reference data point according to the preference degree and the first correction factor and the second correction factor, and obtaining the optimal window length of the reference data point according to the final preference degree includes the following specific steps:
γ′t＝γt×δ1×δ2
Where γ' t represents the final preference level of the t-th initial window of reference data points, γ t represents the preference level of the t-th initial window of reference data points, δ 1 represents a first correction factor for the preference level of the initial window of reference data points, and δ 2 represents a second correction factor for the preference level of the initial window of reference data points;
And then, iteratively increasing the length of an initial window by taking the step length as 8, stopping iteration when the initial window is larger than 100, obtaining the final preference degree of each iterative window, and taking the window length corresponding to the maximum value of the final preference degree as the optimal window length of the t-th reference data point.
Further, the method includes the specific steps of:
Obtaining a box diagram of reference data points according to the determined optimal window length, selecting the number of the data points outside the upper and lower limits in the box diagram as M t, taking the reference data point as a center, taking two data points on the left side and the right side respectively, forming a local range of the reference data point by 5 data points, and then calculating the noise influence degree of the reference data point;
Where σ t denotes the degree of noise influence of the t-th reference data point in the current data, I t denotes the current magnitude of the t-th reference data point, Represents the average current value of all data points within the local range of the t-th reference data point, and norm represents the linear normalization function.
Further, the noise point is removed according to the relation between the noise influence degree and the preset threshold value, and the working state of the battery-powered cabinet is monitored according to the denoised data, including the following specific steps:
For the noise influence degree of each data point in the current data of each power conversion cabinet, when the noise influence degree is larger than a preset threshold T, removing the point, and then interpolating the removed data points through a Lagrange interpolation algorithm to obtain complete denoising data;
when the current value of the battery changing cabinet bin is larger than a preset threshold G, the current at the moment is indicated to be abnormal current, and an alarm is carried out through an alarm system.
The system comprises a memory and a processor, wherein the processor executes a computer program stored in the memory to realize the data online monitoring method for the battery exchange cabinet.
The technical scheme of the invention has the beneficial effects that: the optimal window length, namely the box line diagram length, can be obtained by iterating the data point window length and analyzing the distribution condition of different quantile data in the initial window. Compared with the traditional method for fixing the length of the box diagram, the method for adaptively determining the length of the box diagram can display the abnormal value more prominently according to the actual distribution of the data, is beneficial to the calculation of the influence degree of subsequent noise, and further has a better denoising effect. The optimization degree can be corrected by eliminating the first correction factors of the optimization degree of partial points in each sub-position and by taking the relation between the single-bin current and the total current of the battery-changing cabinet as the second correction factors of the optimization degree and by two correction factors, so that the more accurate and reliable window length is obtained, the calculation of the noise influence degree is more accurate, and the efficiency of the system is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of steps of a method for online monitoring data of a battery exchange cabinet according to the present invention;
fig. 2 is an octal example diagram.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of a specific implementation, structure, characteristics and effects of the data on-line monitoring method and system for a battery-changing cabinet according to the invention in combination with the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a data on-line monitoring method and a system for a battery-changing cabinet.
Referring to fig. 1, a flowchart of a method for online monitoring data of a battery exchange cabinet according to an embodiment of the invention is shown, the method includes the following steps:
s001, collecting current data of the battery changing cabinet;
The invention mainly aims to monitor the current data of the battery exchange cabinet, so that the current data of the battery exchange cabinet needs to be acquired. The power conversion cabinet generally has different scales of 12 bins, 10 bins, 8 bins and the like, in this embodiment, the power conversion cabinet with 8 bins is taken as an example, a current sensor is used to collect the current of each bin and the total current of the power conversion cabinet, the data collection frequency is 50HZ, the collection time is 1 hour, then the current data of each bin and the total current data of the power conversion cabinet are obtained, and the current data of each bin contains a plurality of data points. And then denoising the obtained current data respectively.
S002, calculating the preference degree of the window through the window length of the iterative data points;
it should be noted that, because the box diagram algorithm is extremely sensitive to the distribution of data, the effect of the box diagram algorithm in abnormality detection is completely different for the data with different data distribution characteristics, and further, the accuracy of the calculation of the noise influence degree of each kind of current data also has deviation, so that the denoising effect is affected. Therefore, in this embodiment, the monitored current data is analyzed, the window length of the iterative data points is preset, and the analysis is performed according to the change of the data in the window, so as to obtain the preference degree of the initial window of each data point, and further obtain the optimal window size according to the preference degree, so as to realize the self-adaption of the window length, and then obtain the possible degree that each data point is a noise point according to the self-adaption window. The specific process is as follows:
a. determining a window length and window preference of the current data points;
It should be noted that, when abnormal data is detected by the box map algorithm, the box map only provides a preliminary estimate of the data distribution and outlier condition when the box map is constructed for a smaller window (the number of data points contained in the window is less than 20), and the statistical stability of the box map is poor; while larger windows (containing more than 100 data points within a window) may not provide sufficient data distribution detail due to the excessive amount of data. A preferred window length is therefore between 20 and 100 data points inclusive; therefore, in this embodiment, the initial window length is set to a, a=24 (integer multiple of 8, including 24 data points), and the window selection rule is: taking any data point as a center, recording the any data point as a reference data point, and selecting left and right data points of the reference data point as data contained in an initial window of the data point, wherein the right side of the reference data point contains one more data point than the left side, such as: when the window length is 24, 11 data points are arranged on the left side of the reference data point, and 12 data points are arranged on the right side of the reference data point; in particular, when there are less than eleven data points to the left or right of the reference data point, then the initial window is re-selected centered on the previous or next data point of that data point. The preference degree of the current window is then calculated according to the data change of the window.
Further, because the preference degree of the current window indicates the preference degree with the reference data point as the center, in the process of performing iteration, each iteration is calculated with the preference degree with the reference data point as the center, and when the number of the data points contained in the window is greater than or equal to 100, the iteration is stopped, so that the preference degrees of different window lengths corresponding to the reference data point are obtained. And then carrying out iterative calculation by taking the next data point of the reference data point as a center to obtain the preference degrees of different window lengths corresponding to the next data point of the reference data point, and then sequentially carrying out calculation to obtain the preference degrees of different window lengths corresponding to each data point.
Setting the iteration step length to 8 data points by taking the reference data point as the center, namely adding four data points on the left side and the right side respectively on the basis of the current window. Then calculating window preference degree in each iteration process, wherein the specific implementation method is as follows:
Specifically, the quality of the box line graph construction is mainly dependent on the distribution characteristics of the data in the window, namely, the fewer and the better the number of data points deviating from the median position, the data distribution characteristics can more intuitively identify outliers of the data, and therefore abnormality detection can be performed more accurately. The window data of the reference data points, namely the window containing 24 data points, are taken, the data points in the initial window are sequenced from small to large and recorded as a first data sequence, then the first data sequence is equally divided into eight parts, eight bit intervals are obtained, each bit interval contains three data points, and then the preference degree is obtained by analyzing the data change relations in the bit intervals and among the bit intervals.
As shown in fig. 2, an octal example diagram is shown. For convenience of description, a section from the minimum value to the first octet is hereinafter referred to as a first bit section, a section from the first octet to the second octet is referred to as a second bit section, and so on, to obtain eight bit sections.
Further, firstly, the median of eight bit intervals is obtained, and because the number of data points in each bit interval is the same, the number of data points deviating from the median position is smaller, and the span of data of the bit interval which is closer to the median in the eight bit intervals is smaller, namely the length corresponding to the bit interval; meanwhile, in each bit section, if the data points are more continuous in time sequence, the probability that the data in the section belongs to normal data is higher, which is a relationship in the bit section. The window preference can be characterized by the regularity of span variation between bit intervals and the continuity of data points in each bit interval in time sequence.
Specifically, since the bit interval variation on both sides of the median takes the opposite process, the present embodiment calculates the preference degree of the side smaller than the median (i.e., the minimum to median portion) as an example. Then for the side less than the median, the span should be monotonically decreasing from the first to the fourth bit-wise interval and the more exponentially monotonically decreasing the regularity of the change in the span of the bit-wise interval is. In the embodiment, the average value of the first step difference is used for representing the decreasing relation of the span, the larger the average value is, the more obvious the decreasing relation is, the average value of the second step difference is used for representing the decreasing exponential relation of the span, and the larger the average value is, the more obvious the exponential relation is. The calculation formula of the preference degree is as follows:
Where gamma t denotes the preference degree of the t-th reference data point corresponding to the initial window, gamma 1 denotes the preference degree of the bit section included on the side smaller than the median, Representing a first level difference of the ith partition zone smaller than the median side,/>Representing the second order difference of the ith partition zone smaller than the median side,/>Then the average interval of the corresponding time of the adjacent current data points in the ith bit section on the less-than-median side is represented, y represents the number of bit sections contained on the less-than-median side, exp () represents an exponential function based on a natural constant.
In particular, the method comprises the steps of,The first step of the ith bit zone is represented by the following calculation method: /(I)Wherein L i represents the average value of the current values of the ith bit interval, L i+1 represents the average value of the current values of the (i+1) th bit interval,/>Then the average value of the first step is represented; /(I)Representing the second order difference of the ith bit interval, the calculation method is as follows: /(I)In the method, in the process of the invention,A first step representing the average value of the current value between the (i+1) th bit intervals,/>The average value of the second order difference is represented. /(I)The average interval of the corresponding time of the adjacent current data points in the ith bit interval is represented by the following calculation method: /(I)Where t j denotes a time point corresponding to the first data point, t j+1 denotes a time point corresponding to the (j+1) th data point, and n denotes the number of data points in the (i) th bit interval.
Similarly, a preference degree greater than that on the median data side is obtained:
Wherein gamma 2 represents a degree of preference of a bit section included on the side of the median data, Representing a first level difference of a j-th partition greater than the median side,/>Representing the second order difference of the j-th partition greater than the median side,/>Then the average interval of adjacent current data points corresponding to time in the jth partition on the side of the median is represented.
Then obtaining the preference degree of the initial window of the t-th reference data point according to the obtained preference degree gamma 1 of the bit interval included on the side smaller than the median and the preference degree gamma 2 of the bit interval included on the side larger than the median data
B. Calculating a first correction factor of the degree of preference;
It should be noted that, the preference degree of the data in the calculation window is calculated through the data distribution characteristics in each bit interval, but if there is a point which deviates too much from the overall aggregation trend in a certain bit interval, the calculation of the preference degree is biased, in this embodiment, the time interval with the largest difference between two adjacent data points in all bit intervals is removed, then the average time interval of each bit interval is calculated, the reliability of the window preference degree is represented by the difference between the average time intervals before and after removing, if the difference is smaller, the calculation of the preference degree is more reliable, the correction factor is larger, and the calculation formula of the correction factor is as follows:
Where, δ 1 represents a first correction factor of the preference of the initial window of reference data points, Representing the average interval of adjacent current data points in the ith partition corresponding to time,/>Representing the average interval of the time intervals corresponding to the two largest adjacent current data points removed from the ith partition.
C. Calculating a second correction factor of the window preference degree;
it should be noted that, according to the current conservation theorem, the sum of the current summation of each operation bin and the total monitored current should be approximately equal, so that the difference relationship between the current of each bin and the total current in the same time range can be used to represent the credibility of the current data, so as to be used as the second correction factor of the window preference degree. If the difference value is larger, the possibility that the current moment is influenced by noise is high, larger correction is needed to be carried out on the monitored data point, and a second correction factor calculation formula is as follows:
Wherein delta 2 represents a second correction factor of the initial window preference degree of the reference data point, I i represents total current of the ith data point corresponding to the moment in the window, I i1 represents current value of the 1 st power change cabinet bin which is operated at the ith data point corresponding to the moment, I i2 represents current value of the 2 nd power change cabinet bin which is operated at the ith data point corresponding to the moment, I im represents current value of the m th power change cabinet bin which is operated at the ith data point corresponding to the moment, I im represents current value of the m th bin which is operated at the ith data point corresponding to the moment, m represents number of the power change cabinet bins, N represents window length, and I i-(Ii1+Ii2+…+Iim) represents difference value of the total current and each bin current.
Further, the final preference degree of the initial window of the reference data point is obtained according to the first correction factor and the second correction factor, and the calculation formula is as follows: gamma t′＝γt×δ1×δ2, where gamma t' represents the final preference level of the initial window of reference data points t. And then iterating the reference data points according to a preset iteration step length to obtain the final preference degree of each iteration window, and when the final preference degree obtains the maximum value, representing the window length as the optimal window length of the reference data points. An optimal window length for each data point is then obtained.
S003, obtaining the optimal window length, and calculating the noise influence degree;
For any one data point, after determining the optimal window length, the box diagram can be constructed according to the data points in the window. If the more abnormal points are monitored in the window range of the data point, the more the abnormal points are affected by noise; and the greater the difference between the current value of the data point and the average value of the data in the local range, the greater the noise influence degree is indicated. Taking a reference data point as a center, two data points on the left side and the right side respectively, forming a local range of the reference data point by 5 data points, and then calculating the noise influence degree of the reference data point, wherein the calculation formula is as follows:
Where σ t denotes the degree of noise influence of the t-th reference data point in the current data, I t denotes the current magnitude of the t-th reference data point, The average current value of all data points in the local range of the t reference data point is represented, M t represents the number of data points outside the upper and lower limits, namely the number of abnormal data points, in the box diagram constructed by the data points contained in the optimal window Fan Daxiao, which are centered on the t reference data point, the technology is a known technology, and is not described in detail herein, and norm () represents a linear normalization function.
S004, identifying noise points and denoising according to the noise influence degree;
According to the noise influence degree of each data point in the obtained current data of each power conversion cabinet, when the noise influence degree is larger than a preset threshold value T, the point is considered to be a noise point, the point is removed, T=0.37 is taken as an empirical value, and an implementer can set the value according to different implementation environments. And then interpolation processing is carried out on the removed data points through a Lagrangian difference algorithm, so that complete denoising data is obtained.
And then monitoring the working state of the battery changing cabinet according to the denoised data, and alarming through an alarm system when the current value of the battery changing cabinet bin is larger than a preset threshold G=20 amperes, wherein the current at the moment is the abnormal current.
Through the steps, the online data monitoring method for the battery changing cabinet is completed.
The system comprises a memory and a processor, wherein the processor executes a computer program stored in the memory to realize the data online monitoring method for the battery exchange cabinet.
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 (10)
1. The data on-line monitoring method for the battery exchange cabinet is characterized by comprising the following steps of:
collecting current data of a battery changing cabinet; the power conversion cabinet comprises a plurality of power conversion cabinet bins, the power conversion cabinet current data comprise a plurality of data points, and the data points represent current values collected by each power conversion cabinet bin at each moment;
Marking any one data point as a reference data point, and selecting the initial window length of the reference data point; obtaining a plurality of bit intervals according to the data changes contained in the initial window; obtaining the preference degree of an initial window of the reference data point according to the first-order difference and the second-order difference of the data in the bit interval; obtaining a first correction factor of the preference degree according to the time sequence distribution and the data change of the bit intervals; obtaining a second correction factor of the optimal degree according to the difference value of the current value and the total current value of each battery changing cabinet bin; obtaining final preference degree of the initial window of the reference data point according to the preference degree, the first correction factor and the second correction factor, and obtaining the optimal window length of the reference data point according to the final preference degree;
Constructing a box diagram according to the optimal window length to obtain abnormal data points, obtaining local range data of reference data points, and obtaining noise influence degree of the reference data points according to the variation of current values of the abnormal data points and the local range data; and removing noise points according to the relation between the noise influence degree and a preset threshold value, and monitoring the working state of the battery-replacement cabinet according to the denoised data.
2. The method for online monitoring data of a battery-powered cabinet according to claim 1, wherein the step of marking any one data point as a reference data point and selecting an initial window length of the reference data point comprises the following specific steps:
and constructing an initial window with the length of a by taking the reference data point as the center, wherein the initial window contains a data points, and a is a preset integer value.
3. The method for online monitoring data of a battery-powered cabinet according to claim 1, wherein the obtaining a plurality of bit intervals according to the data changes contained in the initial window comprises the following specific steps:
The data points in the initial window are sequenced from small to large and recorded as a first data sequence, and the first data sequence is equally divided into eight equal divisions to obtain eight bit intervals; the interval from the minimum value to the first octant is recorded as a first quantile interval, the interval from the first octant to the second octant is recorded as a second quantile interval, and the like, so that eight quantile intervals are obtained.
4. The method for online monitoring data of a battery-powered cabinet according to claim 1, wherein the obtaining the preference degree of the initial window of the reference data point according to the first-order difference and the second-order difference of the data in the bit-dividing interval comprises the following specific steps:
respectively obtaining the average value of the current values of all data points in eight bit intervals, arranging the average value of the current values in each bit interval from large to small to obtain a second sequence, and then obtaining the median of the second sequence;
Acquiring a first-order difference and a second-order difference of data points in each bit interval, wherein the first-order difference is obtained by subtracting the average value of the current value of each data point in the first bit interval from the average value of the current value of each data point in the second bit interval, and the average value of the current value of each data point in the first bit interval represents the first-order difference; subtracting the average value of the current value of each data point in the second bit interval from the average value of the current value of each data point in the third bit interval to represent a second first step difference; the second step is a second first step minus the first step, representing the first second step; subtracting the second first step from the third first step represents a second step; the calculation formula for the preference of the initial window of reference data points is as follows:
Where gamma t denotes the preference degree of the t-th reference data point corresponding to the initial window, gamma 1 denotes the preference degree of the bit section included on the side smaller than the median, gamma 2 denotes the preference degree of the bit section included on the side larger than the median, Representing a first level difference of the ith partition zone smaller than the median side,/>Representing the second order difference of the ith partition zone smaller than the median side,/>Represents the average interval of adjacent current data points in the ith partition less than the median side corresponding to time,/>Representing a first level difference of a j-th partition greater than the median side,/>Representing the second order difference of the j-th partition greater than the median side,/>Then the average interval of the corresponding time of the adjacent current data points in the j-th bit interval on the side greater than the median is represented, y represents the number of bit intervals contained on the side less than the median, exp () represents an exponential function based on a natural constant.
5. The method for online monitoring data of a battery-powered cabinet according to claim 1, wherein the obtaining the first correction factor of the preference degree according to the time sequence distribution and the data change of the bit-dividing interval comprises the following specific steps:
Where, δ 1 represents a first correction factor of the preference of the initial window of reference data points, Representing the average interval of adjacent current data points in the ith partition corresponding to time,/>Representing the average interval of the time intervals corresponding to the two largest adjacent current data points removed from the ith partition.
6. The method for online monitoring data of a battery exchange cabinet according to claim 1, wherein the obtaining the second correction factor of the preference degree according to the difference between the current value and the total current value of each battery exchange cabinet bin comprises the following specific steps:
Wherein delta 2 represents a second correction factor of the initial window preference degree of the reference data point, I i represents total current of the ith data point corresponding to the moment in the window, I i1 represents current values of the 1 st power change cabinet bin running at the ith data point corresponding to the moment, I i2 represents current values of the 2 nd power change cabinet bin running at the ith data point corresponding to the moment, I im represents current values of the m th power change cabinet bin running at the ith data point corresponding to the moment, N represents window length, and the total current represents sum of current values of all power change cabinet bins at the corresponding moment, and m represents number of the power change cabinet bins.
7. The method for online monitoring data of a battery-changing cabinet according to claim 1, wherein the obtaining the final preference degree of the initial window of the reference data point according to the preference degree and the first correction factor and the second correction factor, and obtaining the optimal window length of the reference data point according to the final preference degree comprises the following specific steps:
γt ′＝γt×δ1×δ2
Where γ t ′ represents the final preference level of the t-th initial window of reference data points, γ t represents the preference level of the t-th initial window of reference data points, δ 1 represents a first correction factor of the preference level of the initial window of reference data points, and δ 2 represents a second correction factor of the preference level of the initial window of reference data points;
And then, iteratively increasing the length of an initial window by taking the step length as 8, stopping iteration when the initial window is larger than 100, obtaining the final preference degree of each iterative window, and taking the window length corresponding to the maximum value of the final preference degree as the optimal window length of the t-th reference data point.
8. The method for online monitoring data of a battery-changing cabinet according to claim 1, wherein the steps of constructing a box diagram according to the optimal window length to obtain abnormal data points, obtaining local range data of reference data points, and obtaining noise influence degree of the reference data points according to the variation of current values of the abnormal data points and the local range data comprise the following specific steps:
Obtaining a box diagram of reference data points according to the determined optimal window length, selecting the number of the data points outside the upper and lower limits in the box diagram as M t, taking the reference data point as a center, taking two data points on the left side and the right side respectively, forming a local range of the reference data point by 5 data points, and then calculating the noise influence degree of the reference data point;
Where σ t denotes the degree of noise influence of the t-th reference data point in the current data, I t denotes the current magnitude of the t-th reference data point, Represents the average current value of all data points within the local range of the t-th reference data point, and norm represents the linear normalization function.
9. The method for online monitoring data of a battery-changing cabinet according to claim 1, wherein the removing of noise points is performed according to the magnitude relation between the noise influence degree and a preset threshold, and the monitoring of the working state of the battery-changing cabinet is performed according to the denoised data, comprising the following specific steps:
For the noise influence degree of each data point in the current data of each power conversion cabinet, when the noise influence degree is larger than a preset threshold T, removing the point, and then interpolating the removed data points through a Lagrange interpolation algorithm to obtain complete denoising data;
When the current value of the battery changing cabinet bin is larger than a preset threshold G, the current at the moment is abnormal current, and an alarm is given through an alarm system.
10. A data on-line monitoring system for a battery-change cabinet, the system comprising a memory and a processor, wherein the processor executes a computer program stored in the memory, the computer program implementing a data on-line monitoring method for a battery-change cabinet as claimed in any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311836916.6A CN117783745B (en) | 2023-12-28 | 2023-12-28 | Data online monitoring method and system for battery replacement cabinet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311836916.6A CN117783745B (en) | 2023-12-28 | 2023-12-28 | Data online monitoring method and system for battery replacement cabinet |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117783745A CN117783745A (en) | 2024-03-29 |
| CN117783745B true CN117783745B (en) | 2024-06-25 |
Family
ID=90394265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311836916.6A Active CN117783745B (en) | 2023-12-28 | 2023-12-28 | Data online monitoring method and system for battery replacement cabinet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117783745B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117992738B (en) * | 2024-04-03 | 2024-06-11 | 山东兰蝶节能环保科技有限公司 | Automatic electricity-saving monitoring method and system for electric energy of electricity-saving cabinet |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023174002A1 (en) * | 2022-03-18 | 2023-09-21 | 华为技术有限公司 | System monitoring method and apparatus |
| CN117278643A (en) * | 2023-11-20 | 2023-12-22 | 杭州广安汽车电器有限公司 | Vehicle-mounted cloud calibration data transmission system based on cloud edge cooperation |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104537034B (en) * | 2014-12-22 | 2017-11-10 | 国家电网公司 | The Condition Monitoring Data cleaning method of power transmission and transforming equipment based on time series analysis |
| US11341374B2 (en) * | 2018-05-29 | 2022-05-24 | Microsoft Technology Licensing, Llc | Data anomaly detection |
| EP3882922A1 (en) * | 2020-03-21 | 2021-09-22 | Tata Consultancy Services Limited | Discriminating features based sepsis prediction |
| CN111928901A (en) * | 2020-08-19 | 2020-11-13 | 北京兴达智联科技有限公司 | Early warning system and method for power exchange cabinet |
| CN112232447B (en) * | 2020-12-14 | 2021-06-04 | 国网江西省电力有限公司电力科学研究院 | Construction method of complete sample set of power equipment state monitoring data |
| CN113934715B (en) * | 2021-09-25 | 2024-03-01 | 西北工业大学 | Propellant preparation quality data denoising method based on improved non-local mean value |
| CN115145900A (en) * | 2022-06-30 | 2022-10-04 | 南京林业大学 | Big data cleaning method for power battery of electric vehicle |
| CN115347640A (en) * | 2022-08-19 | 2022-11-15 | 深圳供电局有限公司 | Power battery charging abnormity identification early warning method and system and battery replacement station |
| CN115078892B (en) * | 2022-08-19 | 2022-11-01 | 深圳天川电气技术有限公司 | State remote monitoring system for single-machine large-transmission frequency converter |
| CN115238753B (en) * | 2022-09-21 | 2022-12-06 | 西南交通大学 | Self-adaptive SHM data cleaning method based on local outlier factor |
| CN116046187B (en) * | 2023-04-03 | 2023-06-23 | 探长信息技术(苏州)有限公司 | A unusual remote monitoring system of temperature for communication cabinet |
-
2023
- 2023-12-28 CN CN202311836916.6A patent/CN117783745B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023174002A1 (en) * | 2022-03-18 | 2023-09-21 | 华为技术有限公司 | System monitoring method and apparatus |
| CN117278643A (en) * | 2023-11-20 | 2023-12-22 | 杭州广安汽车电器有限公司 | Vehicle-mounted cloud calibration data transmission system based on cloud edge cooperation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117783745A (en) | 2024-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117783745B (en) | Data online monitoring method and system for battery replacement cabinet | |
| WO2020215721A1 (en) | Method for predicting service life of laser radar | |
| CN116502112B (en) | New energy power supply test data management method and system | |
| CN112911627B (en) | Wireless network performance detection method, device and storage medium | |
| CN112433907A (en) | Method and device for processing host operation parameter data of uninterruptible power supply and electronic device | |
| Cai et al. | An estimation model for state of health of lithium-ion batteries using energy-based features | |
| US20220170995A1 (en) | Method and Apparatus for Predicting a State of Health of a Device Battery in a Battery-operated Device | |
| CN112505570A (en) | Method for estimating battery health state of electric automobile | |
| CN117220441A (en) | Information monitoring intelligent early warning method based on panoramic monitoring | |
| CN116643178B (en) | SOC estimation method and related device of battery management system | |
| CN116660759B (en) | Battery life prediction method and device based on BMS battery management system | |
| CN117074965B (en) | Lithium ion battery remaining life prediction method and system | |
| CN115219903A (en) | Battery self-discharge rate abnormity judgment method and device based on Internet of vehicles data analysis | |
| Namdari et al. | An entropy-based approach for modeling lithium-ion battery capacity fade | |
| CN116992311B (en) | Energy storage power supply state analysis method based on machine learning | |
| CN114770607A (en) | Robot health monitoring method and system based on big data | |
| CN113424119A (en) | Work efficiency evaluation method, work efficiency evaluation device, and program | |
| Wang et al. | Large-scale field data-based battery aging prediction driven by statistical features and machine learning | |
| CN117665620A (en) | New energy automobile data-based battery health evaluation method | |
| CN117572241A (en) | Lithium battery fault analysis method based on extended multidimensional time sequence | |
| CN116842464A (en) | Battery system SOC estimation method | |
| US20230305073A1 (en) | Method and apparatus for providing a predicted aging state of a device battery based on a predicted usage pattern | |
| CN117277435A (en) | Health assessment method, system and device for photovoltaic inverter | |
| CN116482540A (en) | Analysis and prediction method, device and system for battery voltage inconsistency | |
| CN115219907A (en) | Lithium battery SOC estimation method, system, medium, equipment and terminal |
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 |