CN117131456B - Multi-sensor data management method for boat bridge electric control system - Google Patents
Multi-sensor data management method for boat bridge electric control system Download PDFInfo
- Publication number
- CN117131456B CN117131456B CN202311382413.6A CN202311382413A CN117131456B CN 117131456 B CN117131456 B CN 117131456B CN 202311382413 A CN202311382413 A CN 202311382413A CN 117131456 B CN117131456 B CN 117131456B
- Authority
- CN
- China
- Prior art keywords
- sensor data
- bridge
- period
- ith
- jth
- 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 36
- 238000013523 data management Methods 0.000 title claims abstract description 21
- 238000012216 screening Methods 0.000 claims abstract description 52
- 230000002159 abnormal effect Effects 0.000 claims abstract description 36
- 230000008859 change Effects 0.000 claims abstract description 7
- 238000001514 detection method Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 9
- CLOMYZFHNHFSIQ-UHFFFAOYSA-N clonixin Chemical compound CC1=C(Cl)C=CC=C1NC1=NC=CC=C1C(O)=O CLOMYZFHNHFSIQ-UHFFFAOYSA-N 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9035—Filtering based on additional data, e.g. user or group profiles
-
- 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
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Abstract
The invention relates to the technical field of sensor data processing, in particular to a multi-sensor data management method for a boat bridge electric control system. The method comprises the following steps: acquiring different sensor data of each bridge in each period in the current time period; obtaining the confusion degree of each sensor data of each bridge to noise according to the change condition of each sensor data of each bridge in each period; obtaining the noise-containing degree of each sensor data of each bridge in each period according to the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion degree of each sensor data of each bridge for noise; and determining screening coefficients of each sensor data of each boat bridge in each period based on the noise level and the local difference, and further screening abnormal sensor data. The invention improves the credibility of the screening result of the abnormal sensor data.
Description
Technical Field
The invention relates to the technical field of sensor data processing, in particular to a multi-sensor data management method for a boat bridge electric control system.
Background
The electric control system for the boat bridge is a system for managing and controlling the boat bridge equipment, and hydraulic pressure, electric control technology and the like are widely applied to the boat bridge equipment, so that quick maneuverability, strong guarantee capability and the like are provided. In practice, when assembled into a bridge, it generally serves as a pass through, such as the passage of support personnel, vehicles and supplies. In order to ensure the safety of traffic, a plurality of sensors arranged on the bridge are monitored through an electric control system, and then the possible abnormal state of the bridge is monitored by combining a local anomaly factor (Local Outlier Factor, LOF) algorithm, so that possible dangers are early-warned and checked in time.
In the prior art, when a person or a vehicle passes through, various parameters of a bridge are monitored through a plurality of sensors, and an abnormal inclination angle or load weight and the like possibly existing in the bridge are monitored through an LOF algorithm, however, in practical application, due to sensor noise, more abnormal data points exist in acquired data, and meanwhile, due to the influence of external environment factors such as water flow and waves, the monitored data fluctuation is often caused to be larger than the actual fluctuation, so that abnormal data and noise data in the monitored data are difficult to distinguish, and when the data are mixed in the rest data, the LOF detection result of the rest data is inaccurate, the data are difficult to monitor, and the reliability of the screening result of the abnormal sensor data of the bridge is lower.
Disclosure of Invention
In order to solve the problem of low reliability of screening results when screening abnormal sensor data of a bridge in the prior art, the invention aims to provide a multi-sensor data management method for a bridge electronic control system, which adopts the following technical scheme:
the invention provides a multi-sensor data management method for a boat bridge electric control system, which comprises the following steps:
acquiring different sensor data of each bridge in each period in the current time period;
obtaining local differences of each sensor data of each bridge and other bridges in each period according to the fluctuation condition of each sensor data of each bridge in each period and the fluctuation condition of the sensor data corresponding to the preset adjacent bridge of each bridge; obtaining the confusion degree of each sensor data of each bridge to noise according to the change condition of each sensor data of each bridge in each period;
obtaining the noise-containing degree of each sensor data of each bridge in each period according to the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion degree of each sensor data of each bridge for noise; determining a screening coefficient of each sensor data of each boat bridge in each period based on the noise level and the local difference;
And screening abnormal sensor data of each bridge in the current time period based on the screening coefficient.
Preferably, the obtaining the local difference between each sensor data of each bridge and the rest of the bridges in each period according to the fluctuation condition of each sensor data of each bridge in each period and the fluctuation condition of the sensor data corresponding to the preset adjacent bridge of each bridge includes:
for the i-th bridge:
respectively calculating the DTW distance between the jth sensor data of the ith bridge in the mth period and the jth sensor data of each preset adjacent bridge in the mth period;
respectively calculating the standard deviation of the jth sensor data of the ith bridge in the mth period and the standard deviation of the jth sensor data of each preset adjacent bridge in the mth period;
obtaining a characteristic coefficient corresponding to each adjacent bridge based on the distance between the ith bridge and each preset adjacent bridge, wherein the distance between the ith bridge and each preset adjacent bridge and the characteristic coefficient are in negative correlation;
and calculating the local difference of the jth sensor data in the mth period between the ith and other boats according to the standard deviation of the jth sensor data in the mth period of the ith boat, the standard deviation of the jth sensor data in the mth period of each preset adjacent boat, the DTW distance and the characteristic coefficient.
Preferably, the local difference of the j-th sensor data between the i-th bridge and the rest of the bridges in the m-th period is calculated by using the following formula:
；
wherein A is imj Representing the local difference of the j-th sensor data in the m-th period between the i-th bridge and the rest of the bridges, d inmj Representing the DTW distance between the jth sensor data of the ith bridge in the mth period and the jth sensor data of the nth preset adjacent bridge of the ith bridge in the mth period,standard deviation epsilon of the jth sensor data of the nth preset adjacent bridge in the mth period is represented by the ith bridge imj Represents the standard deviation, T, of the jth sensor data of the ith bridge in the mth period i,n The characteristic coefficient corresponding to the nth preset adjacent bridge of the ith bridge is represented, N represents the number of the preset adjacent bridges of the ith bridge, norm () represents a normalization function, i represents an absolute value function, and e represents a natural constant.
Preferably, the obtaining the confusion of each sensor data of each bridge to noise according to the change condition of each sensor data of each bridge in each period includes:
for the i-th bridge:
respectively performing curve fitting on all the j-th sensor data of the i-th bridge in each period to obtain a fitting curve of each sensor data of the i-th bridge in each period;
Respectively obtaining the difference between each extreme point on each fitting curve and the average value of all sensor data on the corresponding fitting curve;
taking the ratio of the number of the extreme points on each fitting curve to the number of all the sensor data on the corresponding fitting curve as the number ratio of the extreme points of each fitting curve;
calculating the average time difference between each extreme point and the adjacent extreme point on each fitting curve;
and calculating the confusion degree of the j-th sensor data of the i-th boat bridge on noise according to the difference between each extreme point on each fitting curve and the average value of all the sensor data on the corresponding fitting curve, the number ratio of the extreme points and the average time difference.
Preferably, the confusion of the j-th sensor data of the i-th bridge with respect to noise is calculated using the following formula:
；
wherein B is ij The confusion degree of the jth sensor data of the ith boat bridge to noise is represented, M represents the number of periods in the current time period, G represents the number of extreme points on the fitting curve of the jth sensor data of the ith boat bridge in the mth period, and ρ ijm The number of extreme points of the fitting curve of the jth sensor data of the ith bridge in the mth period is represented as the ratio, deltar ijmg Representing the difference between the g extreme point on the fitting curve of the jth sensor data of the ith bridge in the mth period and the average value of all the sensor data on the fitting curve, deltat ijmg And (e) representing the average time difference between the g extreme point and the adjacent extreme point of the jth sensor data on the fitting curve of the ith bridge in the mth period, wherein e is a natural constant.
Preferably, the obtaining the noise content of each sensor data of each bridge in each period according to the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion degree of each sensor data of each bridge to noise, comprises:
for the ith bridge, the jth sensor data in the mth period:
the difference between the local difference of the j-th sensor data of the ith bridge and the rest of the bridges in the mth period and the local difference of each sensor data except the j-th sensor data of the ith bridge and the rest of the bridges in the mth period is recorded as a first difference corresponding to each sensor data except the j-th sensor data; the sum value between the confusion degree of the jth sensor data of the ith boat bridge and the confusion degree of the ith sensor data except the jth sensor data and the noise is recorded as a first sum value corresponding to each sensor data except the jth sensor data;
And obtaining the noise degree of the jth sensor data of the ith bridge in the mth period according to the first difference value corresponding to each sensor data except the jth sensor data and the first sum value corresponding to each sensor data except the jth sensor data.
Preferably, the noise level of the j-th sensor data of the i-th bridge in the m-th period is calculated by adopting the following formula:
；
wherein Y is imj Indicating the noise level of the jth sensor data of the ith bridge in the mth period, A imj Indicating the local difference of the jth sensor data in the mth period between the ith bridge and the rest of the bridges,representing local differences of the ith and the rest of the sensor data except the jth sensor data in the mth period, J representing the type number of the sensor data, B ij The confusion of the j-th sensor data representing the i-th bridge with respect to noise, ++>Indicating the confusion of the ith bridge and the jth sensor data with respect to noise, norm () indicates the normalization function.
Preferably, the determining a screening coefficient of each sensor data for each bridge in each period based on the noise level and the local difference includes:
For the jth sensor data of the ith bridge in the mth period:
and determining a normalized result of the product of the local difference of the jth sensor data of the ith and other boats and the noise level of the jth sensor data of the ith and other boats and the rest of the jth sensor data in the mth period as a screening coefficient of the jth sensor data of the ith and other boats and the rest of the jth sensor data in the mth period.
Preferably, the screening the abnormal sensor data of each bridge in the current time period based on the screening coefficient includes:
taking the corresponding sensor data in the period of which the screening coefficient is smaller than the preset screening coefficient threshold value as the sensor data to be processed, and performing outlier factor detection on all the sensor data to be processed to obtain discrete sensor data;
and determining the sensor data corresponding to the period when the screening coefficient is greater than or equal to the preset screening coefficient threshold value and all the discrete sensor data as abnormal sensor data.
Preferably, the outlier factor detection is performed on all the sensor data to be processed by adopting a local outlier factor algorithm.
The invention has at least the following beneficial effects:
according to the method, when the traditional method is used for carrying out abnormal detection on different sensor data of the acquired boat bridge, noise can generate certain interference on the acquired sensor data, so that the abnormal sensor data and the normal sensor data are difficult to distinguish, and further accurate screening results of the abnormal sensor data cannot be obtained; the noise-containing degree of each sensor data of each bridge in each period is evaluated according to the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion degree of each sensor data of each bridge to noise; and then, the noise-containing degree and the local difference are combined, the screening coefficient of each sensor data of each bridge in each period is determined, namely, the screening coefficient of each sensor data of each bridge in each period is obtained according to the difference between the bridges and the actual data characteristic analysis, and then, the abnormal sensor data is screened out, so that the screening result of the abnormal sensor data is more accurate and the reliability is higher.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a multi-sensor data management method for a bridge electronic control system according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given to a multi-sensor data management method for a boat bridge electric control system according to the invention by combining the attached drawings and the preferred embodiment.
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 multi-sensor data management method for a boat bridge electric control system, which is concretely described below with reference to the accompanying drawings.
An embodiment of a multi-sensor data management method for a boat bridge electric control system comprises the following steps:
the specific scene aimed at by this embodiment is: when abnormal detection is carried out on sensor data on a bridge, abnormal sensor data and normal sensor data are difficult to distinguish due to water flow, waves, sensor noise and the like, when the LOF algorithm is adopted to detect the collected sensor data, the distribution of the normal sensor data and the abnormal sensor data in an LOF space is mixed together, so that the reliability of screening results of the abnormal sensor data of the bridge is low.
The embodiment provides a multi-sensor data management method for a bridge electronic control system, as shown in fig. 1, and the multi-sensor data management method for the bridge electronic control system of the embodiment comprises the following steps:
Step S1, acquiring different sensor data of each bridge in each period in the current time period.
In this embodiment, first, various sensor data of each bridge in the current time period are acquired through the bridge electronic control system, where the sensor data includes the position, the inclination angle, the load, and the like of the bridge. The current time period in this embodiment is a set formed by all the historical moments with time intervals smaller than or equal to the preset time length from the current moment, and the preset time length in this embodiment is 1 hour, so that the current time period is the last hour, and in specific applications, an implementer can set according to specific situations; in this embodiment, the sensor data is collected every 0.1 seconds, that is, every 0.1 seconds, each sensor data of each bridge is collected, the time interval between adjacent collection moments is 0.1 seconds, and in specific applications, an implementer can set the collection frequency of the sensor data according to specific situations. So far, different sensor data of each collection time of each bridge in the current time period are obtained.
The bridge is used for overcoming the obstacle of the river in military, when a vehicle passes through the bridge, the overall load, the inclination angle and the like of the bridge are correspondingly changed, the specific size is related to the distance between each bridge and the vehicle, in order to ensure the stability of the bridge, the vehicles usually pass through the bridge at uniform speed in sequence at equal intervals when passing through the bridge, therefore, when the vehicles uniformly pass through each bridge, the characteristics are reflected on a plurality of bridges, but the external interference of water flow, waves and the like is reflected only at a certain local position when the noise influence is not considered, but the balance is often broken due to the randomness of the noise, and the possible noise-containing degree in each data is analyzed based on the characteristics, namely, the similarity difference between the randomness of the noise and the range of the data change, so as to achieve the aim of optimizing.
Since the distances between two adjacent vehicles are equal and the speeds of all vehicles on the boat bridge are equal, the ratio between the distance between the two adjacent vehicles and the speed of the vehicles is taken as the length of each period, the current time period is divided to obtain a plurality of sub-time periods, so that the length of each sub-time period after the division and the length of the period are one period, namely, the current time period is divided into a plurality of periods, and it is required to be explained that the length of the last sub-time period may be smaller than the length of the period when the current time period is divided, but the influence on the whole analysis result is not great, and the embodiment also regards the last sub-time period as one period.
So far, by adopting the method, different sensor data of each bridge in each period in the current time period are obtained.
Step S2, obtaining local differences of each sensor data of each bridge and other bridges in each period according to the fluctuation condition of each sensor data of each bridge in each period and the fluctuation condition of the sensor data corresponding to the preset adjacent bridge of each bridge; and obtaining the confusion degree of each sensor data of each bridge to noise according to the change condition of each sensor data of each bridge in each period.
In this embodiment, the current time period is divided to obtain a plurality of periods, and in the same period, the influence of the operation of the vehicle on the whole bridge is relatively close, that is, the differences between the bridges are similar at corresponding moments, so that in this embodiment, the local differences of each sensor data in each period are evaluated according to the fluctuation condition of each sensor data of each bridge in each period and the fluctuation condition of the sensor data corresponding to the preset adjacent bridge of each bridge.
Specifically, for the i-th boat bridge:
in this embodiment, the preset number of the left-hand bridge and the preset number of the right-hand bridge of the ith bridge closest to the left-hand bridge are obtained respectively, the bridge obtained at this time is taken as the preset adjacent bridge of the ith bridge, the preset number in this embodiment is 3, and therefore, the number of the preset adjacent bridges of the ith bridge is 6, and in specific applications, an implementer can set according to specific situations. Then respectively calculating the DTW distance between the jth sensor data of the ith bridge in the mth period and the jth sensor data of each preset adjacent bridge in the mth period; the calculation method of the DTW distance is the prior art, and will not be described in detail here. Then respectively calculating the standard deviation of the jth sensor data of the ith bridge in the mth period and the standard deviation of the jth sensor data of each preset adjacent bridge in the mth period; obtaining a characteristic coefficient corresponding to each adjacent bridge based on the distance between the ith bridge and each preset adjacent bridge, wherein the distance between the ith bridge and each preset adjacent bridge and the characteristic coefficient are in negative correlation; in this embodiment, the distance between the ith bridge and each preset adjacent bridge is obtained, all preset adjacent bridges of the ith bridge are ordered according to the order from smaller distance to larger distance, the preset adjacent bridge sequence of the ith bridge is obtained, each preset adjacent bridge in the preset adjacent bridge sequence is numbered in sequence, the preset adjacent bridges gradually increase from 1, and if the distances between two adjacent preset adjacent bridges in the preset adjacent bridge sequence of the ith bridge and the ith bridge are equal, the numbers of the two preset adjacent bridges are equal, for example: the preset adjacent bridge sequence of the ith bridge is { Q ] i-1 ,Q i+1 ,Q i-2 ,Q i+2 ,Q i-3 ,Q i+3 }, wherein Q i-1 Is the boat bridge with the left side of the ith boat bridge and the nearest distance from the ith boat bridge, Q i+1 Is the boat bridge with the right side of the ith boat bridge and the nearest distance from the ith boat bridge, Q i-2 Is the boat bridge with the left side of the i-1 th boat bridge and the nearest distance from the i-1 th boat bridge, Q i+2 Is the boat bridge with the right side of the ith-1 boat bridge and the nearest distance from the ith-1 boat bridge, Q i-3 Is the boat bridge with the left side of the i-2 th boat bridge and the nearest distance from the i-2 th boat bridge, Q i+3 The boat bridge with the right side of the ith-2 boat bridges being closest to the ith-2 boat bridges; q (Q) i-1 Distance from the ith boat bridge is 10 m, Q i+1 Distance from the ith boat bridge is 10 m, Q i-2 Distance from the ith boat bridge is 20 m, Q i+2 Distance from the ith boat bridge is 20 m, Q i-3 Distance from the ith boat bridge is 30 m, Q i+3 The distance between the boat and the ith boat bridge is 30 meters, Q is i-1 Corresponding reference numerals 1, Q i+1 Corresponding reference numerals 1, Q i-2 Corresponding reference numerals 2, Q i+2 Corresponding reference numerals 2, Q i-3 Corresponding reference numerals 3, Q i+3 The corresponding reference numeral 3. In this embodiment, the nth preset adjacent bridge of the ith bridge is taken as an example for illustration, the method provided in this embodiment may be used to process other preset adjacent bridges of the ith bridge, so as to obtain the feature coefficient corresponding to each preset adjacent bridge of the ith bridge, and for the nth preset adjacent bridge of the ith bridge: taking the natural constant as a base, taking the value of an exponential function with the index number corresponding to the n preset adjacent bridge of the i-th negative bridge as the characteristic coefficient corresponding to the n preset adjacent bridge of the i-th bridge. By adopting the method, the characteristic coefficient corresponding to each preset adjacent bridge of the ith bridge can be obtained.
Further, according to the standard deviation of the jth sensor data of the ith bridge in the mth period, the standard deviation of the jth sensor data of each preset adjacent bridge in the mth period, the DTW distance and the characteristic coefficient, calculating the local difference of the jth sensor data of the ith bridge and the rest of the bridges in the mth period. The specific calculation formula of the local difference of the jth sensor data between the ith bridge and the rest of the bridges in the mth period is as follows:
；
wherein A is imj Representing the local difference of the j-th sensor data in the m-th period between the i-th bridge and the rest of the bridges, d inmj Representing the DTW distance between the jth sensor data of the ith bridge in the mth period and the jth sensor data of the nth preset adjacent bridge of the ith bridge in the mth period,standard deviation epsilon of the jth sensor data of the nth preset adjacent bridge in the mth period is represented by the ith bridge imj Represents the standard deviation, T, of the jth sensor data of the ith bridge in the mth period i,n The characteristic coefficient corresponding to the nth preset adjacent bridge of the ith bridge is represented, N represents the number of the preset adjacent bridges of the ith bridge, norm () represents a normalization function, i represents an absolute value function, and e represents a natural constant.
The smaller the DTW distance between the jth sensor data of the ith bridge in the mth period and the preset adjacent bridge of the ith bridge in the mth period, the higher the similarity between the jth sensor data of the ith bridge in the mth period and the jth sensor data of the preset adjacent bridge in the mth period is. When the standard deviation of the sensor data is larger, that is, the sensor data is generally caused by the reasons of water flow, waves and the like mentioned in the scene, although all the bridges are not interfered by the same water flow at the same time, the local positions, that is, the adjacent bridges are affected, so that when the standard deviation is used as the noise-containing judgment, the embodiment also needs to be adjusted according to the difference between the bridges as the standard deviation, that is, when the standard deviation of the jth sensor data of the nth preset adjacent bridge of the ith bridge in the mth period is larger, and the standard deviation difference between the bridges is also larger, the interference degree of the jth sensor data of the nth preset adjacent bridge in the mth period is relatively higher, and when the similarity analysis is performed on the jth sensor data in the corresponding mth period, the reliability is poorer, that is, the similarity is weakened, that is, amplified, in the situation. The characteristic coefficient corresponding to the preset adjacent bridge of the ith bridge is used for reflecting the influence of the distance on the local difference condition, and the closer the distance is, the larger the corresponding influence should be when the local difference is analyzed, so that the closer the distance between the ith bridge and the preset adjacent bridge is, the larger the characteristic coefficient corresponding to the preset adjacent bridge is. The lower the similarity between the ith and the rest of the bridges, the more the corresponding data features of the bridge do not conform to the features of local similarity, i.e. the higher the degree to which anomalies or noise are represented by the bridge itself, the higher the sensitivity to this data when processed using the LOF algorithm is required.
Different sensors collect different parameter data, and noise is usually obtained by the sensors themselves, so that in the same period, when noise exists in certain parameter data, noise does not represent other parameters, and noise also exists. However, based on further analysis of the characteristics, when the degree of abnormality represented by certain parameter data is large, such as abnormally high load or excessively large inclination angle, the parameters are often related to each other, when certain parameter may have high abnormality, other parameters may also have certain abnormal characteristics, but if the difference between the local difference between certain parameter and other parameters in the same period is large, the local difference may be mainly expressed by noise. Based on this, the present embodiment will obtain the confusion of each sensor data of each bridge for noise according to the variation of each sensor data of each bridge in each period.
For the i-th bridge:
respectively performing curve fitting on all the j-th sensor data of the i-th bridge in each period to obtain a fitting curve of each sensor data of the i-th bridge in each period; the process of curve fitting is prior art and will not be described in detail here. Respectively obtaining the difference between each extreme point on each fitting curve and the average value of all sensor data on the corresponding fitting curve; taking the ratio of the number of the extreme points on each fitting curve to the number of all the sensor data on the corresponding fitting curve as the number ratio of the extreme points of each fitting curve; calculating the average time difference between each extreme point and the adjacent extreme point on each fitting curve; and calculating the confusion degree of the j-th sensor data of the i-th boat bridge on noise according to the difference between each extreme point on each fitting curve and the average value of all the sensor data on the corresponding fitting curve, the number ratio of the extreme points and the average time difference. The specific calculation formula of the confusion degree of the j-th sensor data of the i-th boat bridge to noise is as follows:
；
Wherein B is ij The confusion degree of the jth sensor data of the ith boat bridge to noise is represented, M represents the number of periods in the current time period, G represents the number of extreme points on the fitting curve of the jth sensor data of the ith boat bridge in the mth period, and ρ ijm The number of extreme points of the fitting curve of the jth sensor data of the ith bridge in the mth period is represented as the ratio, deltar ijmg Representing the difference between the g extreme point on the fitting curve of the jth sensor data of the ith bridge in the mth period and the average value of all the sensor data on the fitting curve, deltat ijmg And (e) representing the average time difference between the g extreme point and the adjacent extreme point of the jth sensor data on the fitting curve of the ith bridge in the mth period, wherein e is a natural constant.
The specific acquisition process of the difference between the g extreme point on the fitting curve of the jth sensor data of the ith bridge in the mth period and the average value of all the sensor data on the fitting curve is as follows: and taking the absolute value of the difference between the g extreme point of the ith bridge in the fitting curve of the jth sensor data in the mth period and the average value of all the sensor data in the fitting curve as the difference between the g extreme point of the jth sensor data in the mth period and the average value of all the sensor data in the fitting curve. Since the noise is represented by high frequency, but the collected sensor data itself is rarely represented by high frequency, the extreme point is distributed more densely with the adjacent extreme points The more likely the extreme point is noise at this time, thus byThe characteristic weight is carried out, and the closer the time span is, the smaller the weight is, so that the identification degree of the j-th sensor data to noise obtained according to the extreme point is mainly represented by the data characteristics of the j-th sensor data. The larger the number of extreme points of the fitted curve is, the stronger the fluctuation of the whole parameter data in the period is, so the confidence of the judgment mode based on the data fluctuation on the noise is relatively lower. When the number of extreme points of the fitting curve of the jth sensor data in each period of the ith bridge is larger, the difference between each extreme point of the fitting curve of the jth sensor data in each period of the ith bridge and the mean value of all sensor data in the fitting curve is larger, and the average time difference between each extreme point of the fitting curve of the jth sensor data in each period of the ith bridge and the adjacent extreme point is larger, the jth sensor data is more likely to be fluctuated relative to other parameters, so that the confidence of the judgment mode of noise based on data fluctuation according to standard deviation and the like is relatively lower, namely the confusion degree of the jth sensor data of the ith bridge to noise is larger.
Thus, by adopting the method provided by the embodiment, the local difference of each sensor data of each bridge and the rest of the bridges in each period and the confusion degree of each sensor data of each bridge on noise can be obtained.
Step S3, obtaining the noise-containing degree of each sensor data of each bridge in each period according to the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion degree of each sensor data of each bridge for noise; and determining a screening coefficient of each sensor data of each boat bridge in each period based on the noise level and the local difference.
The present embodiment obtains the local difference of each sensor data for each of the bridge and the remaining bridge in each period and the confusion of each sensor data for each of the bridge with respect to noise in step S2, and then the present embodiment will be based on the difference between the local difference of each sensor data and the local difference of the other sensor data for each of the bridge and the remaining bridge in each period and the confusion of each sensor data for each of the bridge with respect to noise.
Specifically, for the ith bridge, the jth sensor data in the mth period:
the difference between the local difference of the j-th sensor data of the ith bridge and the rest of the bridges in the mth period and the local difference of each sensor data except the j-th sensor data of the ith bridge and the rest of the bridges in the mth period is recorded as a first difference corresponding to each sensor data except the j-th sensor data; it should be noted that: each sensor data except for the jth sensor data corresponds to a first difference. The sum value between the confusion degree of the jth sensor data of the ith boat bridge and the confusion degree of the ith sensor data except the jth sensor data and the noise is recorded as a first sum value corresponding to each sensor data except the jth sensor data; it should be noted that: each sensor data other than the jth sensor data corresponds to a first sum value. And obtaining the noise degree of the jth sensor data of the ith bridge in the mth period according to the first difference value corresponding to each sensor data except the jth sensor data and the first sum value corresponding to each sensor data except the jth sensor data. The specific calculation formula of the noise content of the jth sensor data of the ith bridge in the mth period is as follows:
；
Wherein Y is imj Indicating the noise level of the jth sensor data of the ith bridge in the mth period, A imj Indicating the local difference of the jth sensor data in the mth period between the ith bridge and the rest of the bridges,representing local differences of the ith and the rest of the sensor data except the jth sensor data in the mth period, J representing the type number of the sensor data, B ij The confusion of the j-th sensor data representing the i-th bridge with respect to noise, ++>Indicating the confusion of the ith bridge and the jth sensor data with respect to noise, norm () indicates the normalization function.
The difference between the local difference of the jth sensor data and the local difference of the u sensor data except for the jth sensor data in the mth period of the ith bridge and the rest of the bridges can be reflected, and the larger the difference, the more likely the difference is caused by noise. When the confusion of the jth sensor data of the ith boat bridge with respect to noise and the confusion of the ith sensor data except for the jth sensor data with respect to noise are both high, it is explained that the reliability of the judgment of noise based on the fluctuation difference in the sensor data is poor, and therefore the present embodiment performs the cumulative normalization on the confusion of the two sensor data and subtracts 1, so that the higher the noise confusion, the smaller the weight is given. The degree of noise contained in the jth sensor data in the mth period is greater when the difference between the local difference of the jth sensor data in the mth period and the jth sensor data in the rest of the bridge and the local difference of each sensor data except the jth sensor data in the mth period is greater, the degree of confusion of the jth sensor data of the ith bridge with respect to noise is smaller than the degree of confusion of the jth sensor data except the jth sensor data with respect to noise.
By adopting the method, the noise content of each sensor data of each bridge in each period can be obtained. Next, the present embodiment will determine a screening coefficient of each sensor data for each bridge in each cycle based on the noise level and the local difference.
For the jth sensor data of the ith bridge in the mth period: and determining a normalized result of the product of the local difference of the jth sensor data of the ith and other boats and the noise level of the jth sensor data of the ith and other boats and the rest of the jth sensor data in the mth period as a screening coefficient of the jth sensor data of the ith and other boats and the rest of the jth sensor data in the mth period. In this embodiment, normalization is performed on the product of the local difference of the jth sensor data between the ith bridge and the rest of the bridge in the mth period and the noise level of the jth sensor data between the ith bridge and the rest of the bridge in the mth period by using a norm () normalization function.
So far, by adopting the method provided by the embodiment, the screening coefficient of each sensor data of each bridge in each period can be obtained.
And S4, screening abnormal sensor data of each bridge in each period in the current time period based on the screening coefficient.
The larger the screening coefficient is, the more likely the corresponding sensor data is abnormal sensor data, so that the embodiment initially screens the sensor data based on the screening coefficient to obtain partial abnormal sensor data, and then further detects the rest data by adopting an outlier factor detection algorithm to obtain all abnormal sensor data.
Specifically, sensor data corresponding to a period with a screening coefficient larger than a preset screening coefficient threshold value are used as abnormal sensor data, sensor data corresponding to a period with a screening coefficient smaller than the preset screening coefficient threshold value are used as sensor data to be processed, and an LOF algorithm is adopted to perform outlier factor detection on all the sensor data to be processed to obtain discrete sensor data; and determining the sensor data and all the discrete sensor data corresponding to the period when the screening coefficient is greater than or equal to the preset screening coefficient threshold value as abnormal sensor data, namely obtaining all the abnormal sensor data. The preset screening coefficient threshold in this embodiment is 0.9, and in a specific application, an implementer may set according to a specific situation. The LOF outlier detection algorithm is prior art and will not be described in detail here.
So far, the method provided by the embodiment is adopted to perform anomaly detection on the collected different sensor data in the current time period, and the anomaly sensor data is screened out.
According to the embodiment, when the traditional method is used for carrying out abnormal detection on different sensor data of the acquired boat bridge, noise can generate certain interference on the acquired sensor data, so that the abnormal sensor data and the normal sensor data are difficult to distinguish, and further accurate screening results of the abnormal sensor data cannot be obtained; the noise-containing degree of each sensor data of each bridge in each period is evaluated according to the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion degree of each sensor data of each bridge to noise; and then, the noise-containing degree and the local difference are combined, the screening coefficient of each sensor data of each bridge in each period is determined, namely, the screening coefficient of each sensor data of each bridge in each period is obtained according to the difference between the bridges and the actual data characteristic analysis, and then, the abnormal sensor data is screened out, so that the screening result of the abnormal sensor data is more accurate and the reliability is higher.
It should be noted that: the foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (10)
1. The multi-sensor data management method for the boat bridge electric control system is characterized by comprising the following steps of:
acquiring different sensor data of each bridge in each period in the current time period;
obtaining local differences of each sensor data of each bridge and other bridges in each period according to the fluctuation condition of each sensor data of each bridge in each period and the fluctuation condition of the sensor data corresponding to the preset adjacent bridge of each bridge; obtaining the confusion degree of each sensor data of each bridge to noise according to the change condition of each sensor data of each bridge in each period;
obtaining the noise-containing degree of each sensor data of each bridge in each period according to the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion degree of each sensor data of each bridge for noise; determining a screening coefficient of each sensor data of each boat bridge in each period based on the noise level and the local difference;
And screening abnormal sensor data of each bridge in the current time period based on the screening coefficient.
2. The method for managing multi-sensor data for the electronic control system of the bridge as claimed in claim 1, wherein the obtaining the local difference of each sensor data of each bridge and the rest of the bridges in each period according to the fluctuation of each sensor data of each bridge in each period and the fluctuation of the sensor data corresponding to the preset adjacent bridge of each bridge comprises:
for the i-th bridge:
respectively calculating the DTW distance between the jth sensor data of the ith bridge in the mth period and the jth sensor data of each preset adjacent bridge in the mth period;
respectively calculating the standard deviation of the jth sensor data of the ith bridge in the mth period and the standard deviation of the jth sensor data of each preset adjacent bridge in the mth period;
obtaining a characteristic coefficient corresponding to each adjacent bridge based on the distance between the ith bridge and each preset adjacent bridge, wherein the distance between the ith bridge and each preset adjacent bridge and the characteristic coefficient are in negative correlation;
And calculating the local difference of the jth sensor data in the mth period between the ith and other boats according to the standard deviation of the jth sensor data in the mth period of the ith boat, the standard deviation of the jth sensor data in the mth period of each preset adjacent boat, the DTW distance and the characteristic coefficient.
3. The multi-sensor data management method for the electronic control system of the boat bridge according to claim 2, wherein the local difference of the j-th sensor data of the i-th boat bridge and the rest of the boat bridges in the m-th period is calculated by adopting the following formula:
；
wherein A is imj Representing the local difference of the j-th sensor data in the m-th period between the i-th bridge and the rest of the bridges, d inmj Representing the DTW distance between the jth sensor data of the ith bridge in the mth period and the jth sensor data of the nth preset adjacent bridge of the ith bridge in the mth period,standard deviation epsilon of the jth sensor data of the nth preset adjacent bridge in the mth period is represented by the ith bridge imj Represents the standard deviation, T, of the jth sensor data of the ith bridge in the mth period i,n The characteristic coefficient corresponding to the nth preset adjacent bridge of the ith bridge is represented, N represents the number of the preset adjacent bridges of the ith bridge, norm () represents a normalization function, i represents an absolute value function, and e represents a natural constant.
4. The multi-sensor data management method for the electronic control system of the boat bridge according to claim 1, wherein the obtaining the confusion degree of each sensor data of each boat bridge for noise according to the change condition of each sensor data of each boat bridge in each period comprises the following steps:
for the i-th bridge:
respectively performing curve fitting on all the j-th sensor data of the i-th bridge in each period to obtain a fitting curve of each sensor data of the i-th bridge in each period;
respectively obtaining the difference between each extreme point on each fitting curve and the average value of all sensor data on the corresponding fitting curve;
taking the ratio of the number of the extreme points on each fitting curve to the number of all the sensor data on the corresponding fitting curve as the number ratio of the extreme points of each fitting curve;
calculating the average time difference between each extreme point and the adjacent extreme point on each fitting curve;
and calculating the confusion degree of the j-th sensor data of the i-th boat bridge on noise according to the difference between each extreme point on each fitting curve and the average value of all the sensor data on the corresponding fitting curve, the number ratio of the extreme points and the average time difference.
5. The multi-sensor data management method for the electronic control system of the boat bridge according to claim 4, wherein the confusion degree of the j-th sensor data of the i-th boat bridge to noise is calculated by adopting the following formula:
；
wherein B is ij The confusion degree of the jth sensor data of the ith boat bridge to noise is represented, M represents the number of periods in the current time period, G represents the extreme point on the fitting curve of the jth sensor data of the ith boat bridge in the mth periodNumber ρ of (1) ijm The number of extreme points of the fitting curve of the jth sensor data of the ith bridge in the mth period is represented as the ratio, deltar ijmg Representing the difference between the g extreme point on the fitting curve of the jth sensor data of the ith bridge in the mth period and the average value of all the sensor data on the fitting curve, deltat ijmg And (e) representing the average time difference between the g extreme point and the adjacent extreme point of the jth sensor data on the fitting curve of the ith bridge in the mth period, wherein e is a natural constant.
6. The multi-sensor data management method for a bridge electronic control system according to claim 1, wherein the obtaining the noise level of each sensor data of each bridge in each period based on the difference between the local difference of each sensor data and the local difference of other sensor data of each bridge in each period and the confusion of each sensor data of each bridge with respect to noise, comprises:
For the ith bridge, the jth sensor data in the mth period:
the difference between the local difference of the j-th sensor data of the ith bridge and the rest of the bridges in the mth period and the local difference of each sensor data except the j-th sensor data of the ith bridge and the rest of the bridges in the mth period is recorded as a first difference corresponding to each sensor data except the j-th sensor data; the sum value between the confusion degree of the jth sensor data of the ith boat bridge and the confusion degree of the ith sensor data except the jth sensor data and the noise is recorded as a first sum value corresponding to each sensor data except the jth sensor data;
and obtaining the noise degree of the jth sensor data of the ith bridge in the mth period according to the first difference value corresponding to each sensor data except the jth sensor data and the first sum value corresponding to each sensor data except the jth sensor data.
7. The multi-sensor data management method for the electronic control system of the boat bridge according to claim 6, wherein the noise level of the j-th sensor data of the i-th boat bridge in the m-th period is calculated by adopting the following formula:
；
Wherein Y is imj Indicating the noise level of the jth sensor data of the ith bridge in the mth period, A imj Indicating the local difference of the jth sensor data in the mth period between the ith bridge and the rest of the bridges,representing local differences of the ith and the rest of the sensor data except the jth sensor data in the mth period, J representing the type number of the sensor data, B ij The confusion of the j-th sensor data representing the i-th bridge with respect to noise, ++>Indicating the confusion of the ith bridge and the jth sensor data with respect to noise, norm () indicates the normalization function.
8. The multi-sensor data management method for the electronic control system of the boat bridge according to claim 1, wherein the determining the screening coefficient of each sensor data of each boat bridge in each period based on the noise level and the local difference comprises:
for the jth sensor data of the ith bridge in the mth period:
and determining a normalized result of the product of the local difference of the jth sensor data of the ith and other boats and the noise level of the jth sensor data of the ith and other boats and the rest of the jth sensor data in the mth period as a screening coefficient of the jth sensor data of the ith and other boats and the rest of the jth sensor data in the mth period.
9. The multi-sensor data management method for the electronic control system of the boat bridge according to claim 1, wherein the screening of the abnormal sensor data of each boat bridge in the current time period based on the screening coefficient comprises:
taking the corresponding sensor data in the period of which the screening coefficient is smaller than the preset screening coefficient threshold value as the sensor data to be processed, and performing outlier factor detection on all the sensor data to be processed to obtain discrete sensor data;
and determining the sensor data corresponding to the period when the screening coefficient is greater than or equal to the preset screening coefficient threshold value and all the discrete sensor data as abnormal sensor data.
10. The multi-sensor data management method for the boat bridge electric control system according to claim 9, wherein the outlier factor detection is performed on all the sensor data to be processed by adopting a local outlier factor algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311382413.6A CN117131456B (en) | 2023-10-24 | 2023-10-24 | Multi-sensor data management method for boat bridge electric control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311382413.6A CN117131456B (en) | 2023-10-24 | 2023-10-24 | Multi-sensor data management method for boat bridge electric control system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117131456A CN117131456A (en) | 2023-11-28 |
| CN117131456B true CN117131456B (en) | 2023-12-22 |
Family
ID=88856693
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311382413.6A Active CN117131456B (en) | 2023-10-24 | 2023-10-24 | Multi-sensor data management method for boat bridge electric control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117131456B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107229292A (en) * | 2017-06-29 | 2017-10-03 | 湖北华舟重工应急装备股份有限公司 | A kind of voluntarily bridge of boats electric-control system of multisensor |
| CN116756526A (en) * | 2023-08-17 | 2023-09-15 | 北京英沣特能源技术有限公司 | Full life cycle performance detection and analysis system of energy storage equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ES2928320T3 (en) * | 2018-04-24 | 2022-11-17 | Huebner Gmbh & Co Kg | Passenger boarding bridge with a security device to protect the door of an airplane |
-
2023
- 2023-10-24 CN CN202311382413.6A patent/CN117131456B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107229292A (en) * | 2017-06-29 | 2017-10-03 | 湖北华舟重工应急装备股份有限公司 | A kind of voluntarily bridge of boats electric-control system of multisensor |
| CN116756526A (en) * | 2023-08-17 | 2023-09-15 | 北京英沣特能源技术有限公司 | Full life cycle performance detection and analysis system of energy storage equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117131456A (en) | 2023-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112348237B (en) | Abnormal trend detection method for dynamic drilling data | |
| US10041808B2 (en) | Method of sensor data fusion | |
| CN109325692B (en) | Real-time data analysis method and device for water pipe network | |
| US5357449A (en) | Combining estimates using fuzzy sets | |
| US9207670B2 (en) | Degrading sensor detection implemented within a transmitter | |
| CN116994418B (en) | Pipeline safety early warning method and system | |
| CN111776006A (en) | Pre-alarming method and device for train axle temperature | |
| CN117131456B (en) | Multi-sensor data management method for boat bridge electric control system | |
| JPH08241121A (en) | Print abnormality detecting device | |
| KR102257716B1 (en) | Early warning device and method of power plant using switching technology of multiple-predictive models | |
| CN117390378A (en) | Intelligent management method and system for dual-carbon platform data | |
| KR101883277B1 (en) | Method and device for fault detection of manufacturing process based on dynamic time warping and exponentioal penaltu(dtwep) | |
| CN105222885A (en) | Optical fiber vibration detection method and device | |
| CN116304963A (en) | Data processing system suitable for geological disaster early warning | |
| CN114742103A (en) | City monitoring data processing method and device based on Internet of things and storage medium | |
| CN112083693B (en) | Paint production line information management system based on internet of things technology | |
| CN112258458A (en) | Detection method and system for super high-rise concrete pump pipe based on thermal imaging identification | |
| KR102248536B1 (en) | Early warning device and method of power plant using switching technology of multiple-predictive models | |
| CN116341993B (en) | State monitoring method and system for polystyrene production process | |
| CN114323412B (en) | Water supply pipe network pressure disturbance event detection method | |
| CN117407741A (en) | Abnormal data association analysis early warning method in main and standby dispatching systems | |
| CN116973908A (en) | Target tracking and anti-interference method based on active radar and infrared sensor | |
| CN117854240A (en) | Geological disaster early warning method and geological disaster early warning system | |
| CN117191130A (en) | Multi-scene online temperature and humidity monitoring method and system | |
| CN114323412A (en) | Method for detecting pressure disturbance event of water supply pipe network |
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 |