CN118330376B - Power box with self-diagnosis abnormal electricity utilization module - Google Patents
Power box with self-diagnosis abnormal electricity utilization module Download PDFInfo
- Publication number
- CN118330376B CN118330376B CN202410756704.5A CN202410756704A CN118330376B CN 118330376 B CN118330376 B CN 118330376B CN 202410756704 A CN202410756704 A CN 202410756704A CN 118330376 B CN118330376 B CN 118330376B
- Authority
- CN
- China
- Prior art keywords
- power
- power box
- box
- diagnosis
- self
- 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
- 238000004092 self-diagnosis Methods 0.000 title claims abstract description 164
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 115
- 230000005611 electricity Effects 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 claims abstract description 224
- 230000008569 process Effects 0.000 claims abstract description 214
- 230000005856 abnormality Effects 0.000 claims abstract description 76
- 230000000087 stabilizing effect Effects 0.000 claims abstract description 71
- 238000001514 detection method Methods 0.000 claims abstract description 61
- 230000001965 increasing effect Effects 0.000 claims abstract description 39
- 238000003745 diagnosis Methods 0.000 claims abstract description 31
- 239000012634 fragment Substances 0.000 claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 9
- 230000008859 change Effects 0.000 claims description 18
- 238000009499 grossing Methods 0.000 claims description 18
- 230000004927 fusion Effects 0.000 claims description 16
- 238000012216 screening Methods 0.000 claims description 4
- 230000006641 stabilisation Effects 0.000 claims description 4
- 238000011105 stabilization Methods 0.000 claims description 4
- 230000000739 chaotic effect Effects 0.000 claims description 3
- 230000035945 sensitivity Effects 0.000 abstract description 5
- 230000032683 aging Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000013480 data collection Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000009546 growth abnormality Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The application relates to the technical field of processing of measured electric variables, in particular to a power box with a self-diagnosis abnormal electricity utilization module. Comprising the following steps: and a data acquisition module: collecting power data for the power box; and a data dividing module: dividing the power data into different time lengths; the power box self-diagnosis fragment distinguishing module: dividing the power data difference into a power box power increasing or stabilizing process based on the power data difference in each period of time; and a safety detection module: calculating the power diagnosis warning degree of the power box in the stable process according to the data fluctuation of the growing process and the stable process; an anomaly score calculation module: acquiring power abnormality average scores of the power boxes based on power diagnosis warning degrees of the power boxes and an isolated forest algorithm; and an alarm module: and judging whether the user uses electricity abnormally based on the power box power abnormity average score, and alarming when abnormal electricity is generated. Therefore, the self-diagnosis of the power box is realized, the identification capability of the user on the tiny abnormal power fluctuation is enhanced, and the sensitivity of the abnormal power utilization diagnosis is improved.
Description
Technical Field
The application relates to the technical field of processing of measured electric variables, in particular to a power box with a self-diagnosis abnormal electricity utilization module.
Background
The power box is used as important equipment in the power distribution control link, plays roles in distribution and control in the use process of the electric energy of a user, and is directly related to the normal operation of all electrical equipment in the system. Along with the continuous popularization and wide application of smart power grids, the current power box can collect and count the power consumption data of users besides the functions of power distribution and control, and is provided with a self-diagnosis abnormal power consumption module, and the self-diagnosis abnormal power consumption module is used for helping users monitor the power consumption condition in real time, find abnormal power consumption and take measures in time, so that the energy utilization efficiency is improved, and the energy waste is reduced.
The power box with the self-diagnosis abnormal power consumption module usually detects whether the instantaneous power consumption of a user is stable or not, analyzes the power consumption mode, further judges whether abnormal power consumption conditions such as abnormal increase of electric quantity use or large power fluctuation exist or not, and can only screen instantaneous obvious power data change when detecting the power data of the user by adopting an isolated forest algorithm, the utilization rate of the fine abnormal power fluctuation data is insufficient, the power starting speed is slow and the detection of the fine fluctuation of the power data is not sensitive enough due to electric appliance faults, electric appliance aging and the like, the abnormal power consumption conditions caused by the electric appliance faults and the electric appliance aging cannot be diagnosed, the aged and faulty electric appliances can continuously consume more electric quantity, and risks such as personnel electric shock and fire disaster can also occur, so that the life and property safety of the user are seriously endangered.
Disclosure of Invention
In order to solve the technical problems, the application aims to provide a power box with a self-diagnosis abnormal power utilization module, which adopts the following technical scheme:
The application provides a power box with a self-diagnosis abnormal electricity utilization module, which comprises:
and a data acquisition module: collecting power data of a user power box at each moment;
and a data dividing module: taking a time interval formed by a first preset number of acquisition moments as a self-diagnosis interval of the power box; taking a time segment formed by the self-diagnosis intervals of the second preset number of power boxes as a power box self-diagnosis segment; taking a time window formed by a third preset number of power box self-diagnosis fragments as an abnormal power utilization detection window of the power box;
the power box self-diagnosis fragment distinguishing module: screening effective power box self-diagnosis fragments based on power data difference in each power box self-diagnosis interval, and dividing all the effective power box self-diagnosis fragments into a power box power increasing process and a power box power stabilizing process;
and a safety detection module: obtaining power box power increase abnormality indexes of the power increase processes of the power boxes based on the abnormal fluctuation conditions of the power increase processes of the power boxes; obtaining power box power diagnosis warning degrees of the power stabilizing processes of the power boxes according to the power fluctuation degrees of the power stabilizing processes of the power boxes and the power increase abnormality indexes of the power boxes accumulated before;
an anomaly score calculation module: in the abnormal power utilization detection window of the power box, acquiring power abnormality scores of the power boxes in the power stable process of the power boxes based on power diagnosis warning degrees of the power boxes in the power stable process of the power boxes and an isolated forest algorithm, and acquiring power abnormality average scores of the power boxes in the abnormal power utilization detection window of the power boxes;
And an alarm module: and judging whether the user has abnormal electricity consumption according to the power box power abnormality average score of the power box abnormal electricity consumption detection window, and alarming when abnormal electricity consumption occurs.
In one embodiment, the obtaining process of the self-diagnosis fragment of the effective power box is as follows:
For each power box self-diagnosis interval, calculating the ratio of the maximum power to the minimum power in the power box self-diagnosis interval, and taking the natural constant as logarithm and taking the calculation result of the logarithmic function of which the ratio is true as the interval fluctuation coefficient of the power box self-diagnosis interval; acquiring a sequence consisting of power data at all acquisition moments in a self-diagnosis interval of the power box, and marking the sequence as a first sequence; all elements in the first-order differential sequence of the first sequence and the interval fluctuation coefficient are subjected to weighted fusion to obtain a growth coefficient of a self-diagnosis interval of the power box;
And for each power box self-diagnosis segment, acquiring a third quartile of the growth coefficient of all power box self-diagnosis intervals in the power box self-diagnosis segments, and taking the power box self-diagnosis segment as an effective power box self-diagnosis segment if the third quartile is a non-negative number.
In one embodiment, the distinguishing method of the power box power increasing process and the power box power stabilizing process is as follows:
In each power box self-diagnosis segment, acquiring the average value and the first quartile of the absolute values of the growth coefficients of all the power box self-diagnosis intervals, and defining the power box self-diagnosis segment as a power box power growth process when the average value of the power box self-diagnosis segments is greater than or equal to the first quartile; otherwise, the power box self-diagnostic segment is defined as a power box power smoothing process.
In one embodiment, the power box power increase abnormality index obtaining process of each power box power increase process is:
For each power box power increasing process, marking a sequence consisting of increasing coefficients of all power box self-diagnosis intervals in the power box power increasing process as a second sequence, and taking a forward fusion result of all elements in a first-order differential sequence of the second sequence as an increasing gradient of the power box power increasing process;
obtaining the power range of each power box self-diagnosis interval, and taking the forward fusion result of the power ranges of all the power box self-diagnosis intervals as the change coefficient of the power increase process of the power box;
The power increase abnormality index of the power box in the power increase process of the power box is in positive correlation with the increase gradient and the change coefficient of the power increase process of the power box respectively.
In one embodiment, the obtaining process of the power box power diagnosis warning degree of each power box power stabilizing process is:
obtaining abnormal fluctuation indexes of the power stabilizing process of each power box based on the chaotic degree of data and the abnormal condition of data fluctuation in the self-diagnosis interval of all the power boxes in the power stabilizing process of each power box;
obtaining a comprehensive influence factor of any power box power stable process based on the power box power increase abnormality index of each power box power increase process and the influence degree of each power box power increase process on any power box power stable process;
Acquiring the average value of all power data in the power stabilizing process of each power box, and recording the average value as average power; the power limiting coefficient of the power stabilizing process of each power box is in positive correlation with the average power and in negative correlation with the maximum rated power of the power box;
The power diagnosis warning degree of each power box in the power stabilizing process is in positive correlation with the power limiting coefficient, the comprehensive influence factor and the abnormal fluctuation index of each power box in the power stabilizing process.
In one embodiment, the obtaining process of the abnormal fluctuation index of the power smoothing process of each power box is as follows:
in the power stabilizing process of each power box, smoothing the first sequence of each power box self-diagnosis interval to obtain a power box self-diagnosis power reference sequence of each power box self-diagnosis interval, and calculating the similarity between the first sequence of each power box self-diagnosis interval and the power box self-diagnosis power reference sequence; calculating the confusion of all power data in the self-diagnosis interval of each power box; and the abnormal fluctuation index of the power stabilizing process of the power box is respectively in positive correlation with the similarity and the confusion of the self-diagnosis intervals of each power box in the power stabilizing process of the power box.
In one embodiment, the process of obtaining the comprehensive influence factor of the power smoothing process of any power box is as follows:
In the abnormal electricity utilization detection window of each power box, marking sequence numbers of all power box power increasing processes which occur before any power box power stabilizing process according to time sequence, and forming the sequence numbers into a third sequence of the power stabilizing process of any power box; recording each element in the third sequence as a distance influencing factor;
and taking a forward fusion result of each distance influence factor in the third sequence and the power box power increase abnormality index as a comprehensive influence factor of the power stabilizing process of any power box.
In one embodiment, the process of obtaining the power abnormality score of the power box in the power stabilization process of each power box and obtaining the power abnormality average score of the power box in the power box abnormal electricity utilization detection window is as follows:
In the abnormal power utilization detection window of the power box, the power box power diagnosis warning degree of all the power box power stable processes is input into an isolated forest algorithm, and the power box power abnormal score of each power box power stable process is output; and taking the average value of the power abnormality scores of all the power boxes in the power box abnormal electricity utilization detection window in the power box power stabilizing process as the power box power abnormality average score of the power box abnormal electricity utilization detection window.
In one embodiment, the method for judging whether the user has abnormal electricity consumption is as follows:
Acquiring a fourth preset number of abnormal power utilization detection windows of the power box when normal electric power utilization is performed, and recording the fourth preset number of abnormal power utilization detection windows as normal power utilization detection windows of the power box; recording the power stable process of each power box in the power utilization detection window of each normal power box as a standard power stable process; obtaining power box power abnormality scores of all standard power stable processes according to the power data in all standard power stable processes and the obtaining mode of the power box power abnormality scores;
obtaining an alarm threshold value based on the power abnormality score of the power box in each standard power stable process;
if the average power abnormality score of the power box abnormal power utilization detection window where the latest moment of the power box to be detected is greater than the alarm threshold value, the power utilization of the user is abnormal; otherwise, the user is powered normally.
In one embodiment, the alarm threshold is obtained based on the power abnormality score of the power box in each standard power smoothing process, specifically:
and obtaining the maximum value of the power abnormality scores of the power boxes in all standard power stable processes of any normal power box power utilization detection window, and taking the average value of the maximum values of all normal power box power utilization detection windows as an alarm threshold.
The application has the following beneficial effects:
According to the power change characteristics when the electric appliance is started, calculating the self-diagnosis interval growth coefficient of the electric power box, and improving the sensitivity to power data detection; the utilization efficiency of the user power data is improved by dividing the user power utilization interval into a power box power increasing process and a power box power stabilizing process; calculating power increase abnormality indexes of the power boxes according to the power data of each power box self-diagnosis interval and the change of the increase coefficient of the power box self-diagnosis interval in the power stabilizing process of the power boxes, and enhancing the abnormality diagnosis capability of the power boxes in the power increasing process; calculating the power box power diagnosis warning degree of each power box power stabilizing process through the accumulated power box power increase abnormality index and the difference of power data between the power box self-diagnosis intervals before and after smoothing, amplifying fluctuation abnormality of the power box power data, and improving the accuracy of abnormal electricity diagnosis; the method is characterized in that an isolated forest algorithm is adopted, the input of the algorithm is changed from original user power data to power box power diagnosis warning degree, power box power abnormality average score is obtained, power fluctuation abnormality scores in the aging and fault processes of the electric appliance are accumulated, the identification capability of fine abnormal power fluctuation of the user is enhanced, and the sensitivity of abnormal power diagnosis is improved.
Drawings
In order to more clearly illustrate the embodiments of the application 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 application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of a power box with a self-diagnostic abnormal electricity usage module according to one embodiment of the present application;
FIG. 2 is a schematic flow diagram of a power box with a self-diagnostic abnormal electricity usage module.
Detailed Description
In order to further describe the technical means and effects adopted by the application to achieve the preset aim, the following is a detailed description of the specific implementation, structure, characteristics and effects of the power box with the self-diagnosis abnormal electricity utilization module according to the application with reference to the attached 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 application belongs.
The following specifically describes a specific scheme of a power box with a self-diagnosis abnormal electricity utilization module provided by the application with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of a power box with a self-diagnosing abnormal electricity usage module according to an embodiment of the present application is shown, including:
the data acquisition module 101: and collecting the power consumption data of the user power box at each moment.
As an example, the present step may include the steps of:
And collecting the power consumption data of the user power box at each moment through a power meter. Preferably, the data acquisition time interval in this embodiment is 0.1 seconds. It should be noted that, the data collection time interval and the data collection duration may be set by the user, and the embodiment is not limited specifically.
The data dividing module 102: dividing a power box self-diagnosis interval, a power box self-diagnosis fragment and a power box abnormal electricity utilization detection window, wherein the power box self-diagnosis section and the power box abnormal electricity utilization detection window specifically comprise:
Taking a time interval formed by a first preset number of connected acquisition moments as a power box self-diagnosis interval, and dividing the power box self-diagnosis interval from the latest moment onwards to obtain each power box self-diagnosis interval; taking a time segment formed by a second preset number of connected power box self-diagnosis intervals as a power box self-diagnosis segment, and dividing the power box self-diagnosis segments forwards from the latest moment to obtain each power box self-diagnosis segment; and taking a time window formed by a third preset number of connected power box self-diagnosis fragments as a power box abnormal electricity utilization detection window, and dividing the power box abnormal electricity utilization detection window forward from the latest moment to obtain each power box abnormal electricity utilization detection window.
It should be noted that the first, second and third preset numbers can be set by the practitioner, and in this embodiment, the first, second and third preset numbers are set to 50, 30 and 500, respectively.
Power box self-diagnostic segment discrimination module 103: and screening effective power box self-diagnosis fragments based on the power data difference in each power box self-diagnosis interval, and dividing all the effective power box self-diagnosis fragments into a power box power increasing process and a power box power stabilizing process.
As an example, the present step may include the steps of:
(1) The growth coefficient of each power box self-diagnosis interval is obtained based on the power difference, the maximum power and the minimum power of adjacent acquisition time in each power box self-diagnosis interval, and specifically comprises the following steps:
Home appliances usually require a short time to reach steady state at start-up, which depends on the characteristics of the appliance itself and the power requirements. A normal home appliance can reach a stable operation state in a few seconds and the power rise speed is high. The power supply of the electrical appliance is unstable due to the performance reduction of the internal electrical appliance elements, the electrical appliance can reach a stable state only after a long time, the power rising speed of the electrical appliance is low when the electrical appliance is started, and the rising speed fluctuates.
For each power box self-diagnosis interval, calculating the ratio of the maximum power to the minimum power in the power box self-diagnosis interval, and taking the natural constant as logarithm and taking the calculation result of the logarithmic function of which the ratio is true as the interval fluctuation coefficient of the power box self-diagnosis interval; acquiring a sequence consisting of power data at all acquisition moments in a self-diagnosis interval of the power box, and marking the sequence as a first sequence; and carrying out weighted fusion on all elements in the first-order differential sequence of the first sequence and the interval fluctuation coefficient to obtain the growth coefficient of the self-diagnosis interval of the power box.
Preferably, as an embodiment of the present application, a product of a sum value of all elements in the first-order differential sequence of the first sequence and the interval fluctuation coefficient may be used as a growth coefficient of the power box self-diagnosis interval.
When a user turns on the electric appliance, the power data is continuously increased, the increase coefficient of the self-diagnosis interval of the electric power box is positive, and when the phase difference of adjacent power data in the self-diagnosis interval of the electric power box is larger, the power increase in the self-diagnosis interval of the electric power box is faster, and the reflected increase coefficient of the self-diagnosis interval of the electric power box is larger. When the user turns off the electric appliance, the power data is rapidly reduced, the self-diagnosis interval growth coefficient of the power box is negative, and the absolute value of the self-diagnosis interval growth coefficient is larger. When the electric appliance is normally used, the change of the power data is not large, and the absolute value of the self-diagnosis interval growth coefficient of the power box is smaller.
(2) Screening effective power box self-diagnosis fragments, specifically:
As the power variation when the fault electric appliance and the normal electric appliance are closed is not greatly different, for each power box self-diagnosis segment, obtaining a third quartile of the growth coefficient of all power box self-diagnosis intervals in the power box self-diagnosis segments, and discarding the power box self-diagnosis segment if the third quartile is negative; and if the third quartile is a non-negative number, taking the power box self-diagnosis fragment as an effective power box self-diagnosis fragment.
(3) The power box power increasing process and the power box power stabilizing process are distinguished for each effective power box self-diagnosis segment, and the method specifically comprises the following steps:
in each power box self-diagnosis segment, acquiring the average value and the first quartile of the absolute values of the growth coefficients of all the power box self-diagnosis intervals, and defining the power box self-diagnosis segment as a power box power growth process when the average value of the power box self-diagnosis segment is greater than or equal to the first quartile; when the average value of the power box self-diagnosis fragments is smaller than the first quartile, the power box self-diagnosis fragments are defined as a power box power smoothing process.
The security detection module 104: obtaining power box power increase abnormality indexes of the power increase processes of the power boxes based on the abnormal fluctuation conditions of the power increase processes of the power boxes; and obtaining the power box power diagnosis warning degree of each power box power stabilizing process according to the power fluctuation degree of each power box power stabilizing process and the power increase abnormality index of the power box accumulated before.
(1) When each power box self-diagnosis segment is a power box power increasing process, obtaining a power box power increasing abnormality index of each power box power increasing process based on the increasing coefficients of two adjacent power box self-diagnosis intervals and the range of each power box self-diagnosis interval, specifically:
(1.1) obtaining a growth gradient of each power increase process of the power boxes based on growth coefficients of the self-diagnosis intervals of the two adjacent power boxes:
And for each power box power increasing process, marking a sequence consisting of increasing coefficients of all power box self-diagnosis intervals in the power box power increasing process as a second sequence, and taking a forward fusion result of all elements in a first-order differential sequence of the second sequence as an increasing gradient of the power box power increasing process.
Preferably, as an embodiment of the present application, the forward fusion may be summing the normalized values of all the elements in the first-order differential sequence of the second sequence, and as other embodiments, on the basis of achieving the purpose of forward fusion of all the elements, an implementer may obtain the forward fusion result of all the elements by using other methods, which is not limited in particular.
(1.2) Obtaining a change coefficient of the power increase process of each power box based on the power range of the self-diagnosis interval of each power box:
And for each power box power increasing process, acquiring the power range of each power box self-diagnosis interval, and taking the forward fusion result of the power ranges of all the power box self-diagnosis intervals as the change coefficient of the power box power increasing process.
Preferably, as an embodiment of the present application, the forward fusion of the power limit may be a sum of the power limits of all power box self-diagnosis intervals.
(1.3) Obtaining power box power increase abnormality indexes of each power box power increase process based on the increase gradient and the change coefficient:
the power increase abnormality index of each power box in the power increase process is in positive correlation with the increase gradient and the change coefficient of each power box in the power increase process.
Preferably, as one embodiment of the present application, an exponential function based on the inverse of a natural constant and based on the change coefficient of each power tank power increase process as an exponent is recorded as a first exponential function; and taking the product of the calculation result of the first exponential function and the growth gradient of each power box power growth process as the power box power growth abnormality exponent of each power box power growth process.
When the power increase rate of the power box power increase process is slower, the starting abnormality of the electric appliance is more obvious, the change coefficient is smaller, the first exponential function is larger, and the corresponding power box power increase abnormality index is larger. In addition, when the electrical appliance is aged or fails, the longer the duration of the opening process is, the more unstable the power increase change is, the longer the duration of the power box self-diagnosis interval increase coefficient reaches the vicinity of zero value in the reflected power box power increase process is, the larger the difference between the corresponding accumulated adjacent power box self-diagnosis interval increase coefficients is, namely the larger the increase gradient is, and the larger the power box power increase abnormality index is.
(2) When each power box self-diagnosis segment is a power box power stable process, obtaining an abnormal fluctuation index of each power box power stable process based on the chaotic degree of data and the abnormal condition of data fluctuation in all power box self-diagnosis intervals of each power box power stable process, specifically:
in the power stabilizing process of each power box, smoothing the first sequence of each power box self-diagnosis interval to obtain a power box self-diagnosis power reference sequence of each power box self-diagnosis interval, and calculating the similarity between the first sequence of each power box self-diagnosis interval and the power box self-diagnosis power reference sequence; calculating the confusion of all power data in the self-diagnosis interval of each power box; and the abnormal fluctuation index of the power stabilizing process of the power box is respectively in positive correlation with the similarity and the confusion of the self-diagnosis intervals of each power box in the power stabilizing process of the power box.
Preferably, as an embodiment of the present application, a product of the similarity and the confusion of each power box self-diagnosis section is calculated and recorded as a first product; the abnormal fluctuation index of the power tank power smoothing process may be a sum of the first products of all power tank self-diagnosis intervals in the power tank power smoothing process. The degree of confusion can be the information entropy of all data in each power box self-diagnosis interval.
(3) In the abnormal electricity utilization detection window of each power box, based on the power box power increase abnormality index of each power box power increase process and the influence degree of each power box power increase process on any power box power stable process, the comprehensive influence factor of any power box power stable process is obtained, specifically:
(3.1) obtaining a distance influence factor of each power box power increasing process on any power box power stabilizing process:
In the abnormal electricity utilization detection window of each power box, marking sequence numbers of all power box power increasing processes which occur before any power box power stabilizing process according to time sequence, and forming the sequence numbers into a third sequence of the power stabilizing process of any power box; and recording each element in the third sequence as a distance influencing factor.
It should be noted that, the embodiment of the present application only provides a method for obtaining a distance influencing factor, and as other embodiments, the distance influencing factor may be obtained by an implementer by other methods based on the purpose of obtaining the distance influencing factor of each power box power increasing process to any power box power stabilizing process based on the distance of each power box power increasing process to any power box power stabilizing process.
(3.2) Taking the forward fusion result of the distance influence factor of each power box power increase process and the power box power increase abnormality index in the third sequence as the comprehensive influence factor of any power box power stabilizing process.
Preferably, as an embodiment of the present application, a product of a square of a distance influence factor of each power box power increase process in the third sequence and an abnormality index of power box power increase is calculated and recorded as a second product; the integrated impact factor of any one of the power tank power smoothing processes may be a sum of the second products of all power tank power increase processes in the third sequence.
(4) Obtaining a power limiting coefficient of each power box power stabilizing process according to the power magnitude of each power box power stabilizing process:
Acquiring the average value of all power data in the power stabilizing process of each power box, and recording the average value as average power; the power limiting coefficient of the power stabilizing process of each power box is in positive correlation with the average power and in negative correlation with the maximum rated power of the power box.
Preferably, as one embodiment of the present application, a ratio of the average power of the power smoothing process of each power box to the maximum rated power of the power box is calculated and is recorded as a first ratio; and (3) taking a calculation result of an exponential function taking a natural constant as a base and taking the first ratio as an index as a power limiting coefficient of the power stabilizing process of each power box.
(5) Obtaining power box power diagnosis warning degree of each power box power stable process based on power limiting coefficient, comprehensive influence factor and abnormal fluctuation index of each power box power stable process:
The power diagnosis warning degree of each power box in the power stabilizing process is in positive correlation with the power limiting coefficient, the comprehensive influence factor and the abnormal fluctuation index of each power box in the power stabilizing process.
Preferably, as an embodiment of the present application, the power box power diagnostic warning of each power box power smoothing process may be a product of a power limiting coefficient, a comprehensive influence factor, and an abnormal fluctuation index of each power box power smoothing process.
The power box power diagnosis warning degree of each power box power stabilizing process in the power box abnormal power utilization detection window can be calculated through the process. When the power increase abnormality index of the power box in the power increase process before the power stabilizing process of the power box is larger, the abnormality when the electric appliance is started is more obvious, the possibility of power fluctuation abnormality in the power stabilizing process of the corresponding power box is higher, the comprehensive influence factor is larger, and the calculated power diagnosis warning degree of the power box is larger; meanwhile, in the power increase process of the power box, which is closer to the occurrence time of the current power box power stabilization process, the influence of the corresponding electric appliance on power fluctuation of the power box is larger, so that the power increase abnormality index of the power box, which is closer to the occurrence time of the current power box power stabilization process, is given a higher weight. In addition, when the power detected by the power box is higher, the number of opened electric appliances is larger, and abnormal power fluctuation can be covered in normal power change, so that the power box power diagnosis warning degree corresponding to the power stabilizing process of the power box with higher average power is improved by weighting the power limiting coefficient. After the appliance enters an operating state, its power is typically maintained near the rated power. When an electrical appliance is aged or fails, the power of the electrical appliance fluctuates near rated power, and the smaller the similarity between a corresponding power box self-diagnosis interval and a smoothed power box self-diagnosis power reference sequence is, the larger the Euclidean distance between the two sequences is; meanwhile, when the power data of the power box fluctuates, the larger the information entropy of the self-diagnosis interval of the power box is, the larger the fluctuation index is, and the larger the power diagnosis warning degree of the power box corresponding to the power stabilizing process of the power box is.
The anomaly score calculation module 105: in the abnormal power utilization detection window of the power box, based on the power diagnosis warning degree of the power box in the power stable process of each power box, the power abnormality score of the power box in the power stable process of each power box is obtained by utilizing an isolated forest algorithm, and then the power abnormality average score of the abnormal power utilization detection window of the power box is obtained.
In the power box abnormal electricity utilization detection window, the power box power diagnosis warning degree of all power box power stabilizing processes is taken as a total sample, the number of the isolated trees is set to be 100, the number of sub-samples extracted each time is set to be 256, the maximum depth of the isolated trees is set to be 8, and it is noted that the number of the isolated trees, the number of sub-samples extracted each time and the maximum depth of the isolated trees can be set by an implementer per se. And taking the average value of the power abnormality scores of all the power boxes in the power box abnormal electricity utilization detection window in the power box power stabilizing process as the power box power abnormality average score of the power box abnormal electricity utilization detection window.
Alarm module 106: and judging whether the user has abnormal electricity consumption according to the power box power abnormality average score of the power box abnormal electricity consumption detection window, and alarming when abnormal electricity consumption occurs.
The abnormal electricity utilization detection windows of the fourth preset number of the electric boxes when the normal electricity utilization is obtained are recorded as the normal electricity utilization detection windows of the electric boxes, and it is to be noted that the fourth preset number of the electric boxes can be set by an implementer, and the fourth preset number of the electric boxes is set to be 500 in the embodiment.
Recording the power stable process of each power box in the power utilization detection window of each normal power box as a standard power stable process; and obtaining the power abnormality score of the power box in each standard power stable process according to the power data in each standard power stable process and the obtaining mode of the power abnormality score of the power box. And obtaining the maximum value of the power abnormality scores of the power boxes in all standard power stable processes of any normal power box power utilization detection window, and taking the average value of the maximum values of all normal power box power utilization detection windows as an alarm threshold.
If the average power abnormality score of the power box abnormal electricity utilization detection window where the latest moment of the power box to be detected is greater than the alarm threshold value, the user uses electricity abnormally and alarms; otherwise, the user is powered normally and does not alarm.
When the user electrical appliance is aged or fails, the power abnormality scores of the power boxes in the power box abnormal electricity utilization detection window are accumulated continuously, and when the electrical appliance is aged or fails seriously, the calculated average score of the power boxes exceeds an alarm threshold value. At this time, the safety detection module transmits an alarm signal to the alarm module, alarms abnormal electricity consumption, and completes self-diagnosis of the abnormal electricity consumption.
The schematic flow diagram of the power box with the self-diagnosis abnormal power utilization module is shown in fig. 2.
In summary, according to the embodiment of the application, the self-diagnosis interval growth coefficient of the power box is calculated according to the power change characteristic when the electric appliance is started, so that the sensitivity to power data detection is improved; the utilization efficiency of the user power data is improved by dividing the user power utilization interval into a power box power increasing process and a power box power stabilizing process; calculating power increase abnormality indexes of the power boxes according to the power data of each power box self-diagnosis interval and the change of the increase coefficient of the power box self-diagnosis interval in the power stabilizing process of the power boxes, and enhancing the abnormality diagnosis capability of the power boxes in the power increasing process; calculating the power box power diagnosis warning degree of each power box power stabilizing process through the accumulated power box power increase abnormality index and the difference of power data between the power box self-diagnosis intervals before and after smoothing, amplifying fluctuation abnormality of the power box power data, and improving the accuracy of abnormal electricity diagnosis; the method is characterized in that an isolated forest algorithm is adopted, the input of the algorithm is changed from original user power data to power box power diagnosis warning degree, power box power abnormality average score is obtained, power fluctuation abnormality scores in the aging and fault processes of the electric appliance are accumulated, the identification capability of fine abnormal power fluctuation of the user is enhanced, and the sensitivity of abnormal power diagnosis is improved.
It should be noted that: the sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and not for limiting the same; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.
Claims (10)
1. A power box with self-diagnosing abnormal electricity usage module, comprising:
and a data acquisition module: collecting power data of a user power box at each moment;
and a data dividing module: taking a time interval formed by a first preset number of acquisition moments as a self-diagnosis interval of the power box; taking a time segment formed by the self-diagnosis intervals of the second preset number of power boxes as a power box self-diagnosis segment; taking a time window formed by a third preset number of power box self-diagnosis fragments as an abnormal power utilization detection window of the power box;
the power box self-diagnosis fragment distinguishing module: screening effective power box self-diagnosis fragments based on power data difference in each power box self-diagnosis interval, and dividing all the effective power box self-diagnosis fragments into a power box power increasing process and a power box power stabilizing process;
and a safety detection module: obtaining power box power increase abnormality indexes of the power increase processes of the power boxes based on the abnormal fluctuation conditions of the power increase processes of the power boxes; obtaining power box power diagnosis warning degrees of the power stabilizing processes of the power boxes according to the power fluctuation degrees of the power stabilizing processes of the power boxes and the power increase abnormality indexes of the power boxes accumulated before;
an anomaly score calculation module: in the abnormal power utilization detection window of the power box, acquiring power abnormality scores of the power boxes in the power stable process of the power boxes based on power diagnosis warning degrees of the power boxes in the power stable process of the power boxes and an isolated forest algorithm, and acquiring power abnormality average scores of the power boxes in the abnormal power utilization detection window of the power boxes;
And an alarm module: and judging whether the user has abnormal electricity consumption according to the power box power abnormality average score of the power box abnormal electricity consumption detection window, and alarming when abnormal electricity consumption occurs.
2. The power box with self-diagnostic abnormal electricity usage module according to claim 1, wherein the obtaining process of the effective power box self-diagnostic fragment is:
For each power box self-diagnosis interval, calculating the ratio of the maximum power to the minimum power in the power box self-diagnosis interval, and taking the natural constant as logarithm and taking the calculation result of the logarithmic function of which the ratio is true as the interval fluctuation coefficient of the power box self-diagnosis interval; acquiring a sequence consisting of power data at all acquisition moments in a self-diagnosis interval of the power box, and marking the sequence as a first sequence; all elements in the first-order differential sequence of the first sequence and the interval fluctuation coefficient are subjected to weighted fusion to obtain a growth coefficient of a self-diagnosis interval of the power box;
And for each power box self-diagnosis segment, acquiring a third quartile of the growth coefficient of all power box self-diagnosis intervals in the power box self-diagnosis segments, and taking the power box self-diagnosis segment as an effective power box self-diagnosis segment if the third quartile is a non-negative number.
3. The power box with self-diagnosing abnormal electricity usage module according to claim 1, wherein the distinguishing method of the power box power increasing process and the power box power smoothing process is as follows:
In each power box self-diagnosis segment, acquiring the average value and the first quartile of the absolute values of the growth coefficients of all the power box self-diagnosis intervals, and defining the power box self-diagnosis segment as a power box power growth process when the average value of the power box self-diagnosis segments is greater than or equal to the first quartile; otherwise, the power box self-diagnostic segment is defined as a power box power smoothing process.
4. The power box with self-diagnosing abnormal power consumption module as recited in claim 1, wherein the power box power increase abnormality index obtaining process of each power box power increase process is as follows:
For each power box power increasing process, marking a sequence consisting of increasing coefficients of all power box self-diagnosis intervals in the power box power increasing process as a second sequence, and taking a forward fusion result of all elements in a first-order differential sequence of the second sequence as an increasing gradient of the power box power increasing process;
obtaining the power range of each power box self-diagnosis interval, and taking the forward fusion result of the power ranges of all the power box self-diagnosis intervals as the change coefficient of the power increase process of the power box;
The power increase abnormality index of the power box in the power increase process of the power box is in positive correlation with the increase gradient and the change coefficient of the power increase process of the power box respectively.
5. The power box with self-diagnosing abnormal electricity consumption module according to claim 2, wherein the power box power diagnostic warning degree of each power box power stabilizing process is obtained by the following steps:
obtaining abnormal fluctuation indexes of the power stabilizing process of each power box based on the chaotic degree of data and the abnormal condition of data fluctuation in the self-diagnosis interval of all the power boxes in the power stabilizing process of each power box;
obtaining a comprehensive influence factor of any power box power stable process based on the power box power increase abnormality index of each power box power increase process and the influence degree of each power box power increase process on any power box power stable process;
Acquiring the average value of all power data in the power stabilizing process of each power box, and recording the average value as average power; the power limiting coefficient of the power stabilizing process of each power box is in positive correlation with the average power and in negative correlation with the maximum rated power of the power box;
The power diagnosis warning degree of each power box in the power stabilizing process is in positive correlation with the power limiting coefficient, the comprehensive influence factor and the abnormal fluctuation index of each power box in the power stabilizing process.
6. The power box with the self-diagnosis abnormal power consumption module according to claim 5, wherein the obtaining process of the abnormal fluctuation index of the power stabilizing process of each power box is as follows:
in the power stabilizing process of each power box, smoothing the first sequence of each power box self-diagnosis interval to obtain a power box self-diagnosis power reference sequence of each power box self-diagnosis interval, and calculating the similarity between the first sequence of each power box self-diagnosis interval and the power box self-diagnosis power reference sequence; calculating the confusion of all power data in the self-diagnosis interval of each power box; and the abnormal fluctuation index of the power stabilizing process of the power box is respectively in positive correlation with the similarity and the confusion of the self-diagnosis intervals of each power box in the power stabilizing process of the power box.
7. The power box with self-diagnosis abnormal electricity utilization module according to claim 5, wherein the process of obtaining the comprehensive influence factor of the power stabilizing process of any power box is as follows:
In the abnormal electricity utilization detection window of each power box, marking sequence numbers of all power box power increasing processes which occur before any power box power stabilizing process according to time sequence, and forming the sequence numbers into a third sequence of the power stabilizing process of any power box; recording each element in the third sequence as a distance influencing factor;
and taking a forward fusion result of each distance influence factor in the third sequence and the power box power increase abnormality index as a comprehensive influence factor of the power stabilizing process of any power box.
8. The power box with the self-diagnosis abnormal power consumption module according to claim 1, wherein the process of obtaining the power box power abnormality score of each power box power stabilization process and obtaining the power box power abnormality average score of the power box abnormal power consumption detection window is as follows:
In the abnormal power utilization detection window of the power box, the power box power diagnosis warning degree of all the power box power stable processes is input into an isolated forest algorithm, and the power box power abnormal score of each power box power stable process is output; and taking the average value of the power abnormality scores of all the power boxes in the power box abnormal electricity utilization detection window in the power box power stabilizing process as the power box power abnormality average score of the power box abnormal electricity utilization detection window.
9. The power box with self-diagnosis abnormal electricity utilization module according to claim 1, wherein the method for judging whether the user has abnormal electricity utilization is as follows:
Acquiring a fourth preset number of abnormal power utilization detection windows of the power box when normal electric power utilization is performed, and recording the fourth preset number of abnormal power utilization detection windows as normal power utilization detection windows of the power box; recording the power stable process of each power box in the power utilization detection window of each normal power box as a standard power stable process; obtaining power box power abnormality scores of all standard power stable processes according to the power data in all standard power stable processes and the obtaining mode of the power box power abnormality scores;
obtaining an alarm threshold value based on the power abnormality score of the power box in each standard power stable process;
if the average power abnormality score of the power box abnormal power utilization detection window where the latest moment of the power box to be detected is greater than the alarm threshold value, the power utilization of the user is abnormal; otherwise, the user is powered normally.
10. The power box with the self-diagnosis abnormal power consumption module according to claim 9, wherein the power box power abnormality score based on each standard power plateau process obtains an alarm threshold value, specifically:
and obtaining the maximum value of the power abnormality scores of the power boxes in all standard power stable processes of any normal power box power utilization detection window, and taking the average value of the maximum values of all normal power box power utilization detection windows as an alarm threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410756704.5A CN118330376B (en) | 2024-06-13 | 2024-06-13 | Power box with self-diagnosis abnormal electricity utilization module |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410756704.5A CN118330376B (en) | 2024-06-13 | 2024-06-13 | Power box with self-diagnosis abnormal electricity utilization module |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118330376A CN118330376A (en) | 2024-07-12 |
| CN118330376B true CN118330376B (en) | 2024-08-09 |
Family
ID=91779236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410756704.5A Active CN118330376B (en) | 2024-06-13 | 2024-06-13 | Power box with self-diagnosis abnormal electricity utilization module |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118330376B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109239494A (en) * | 2018-09-21 | 2019-01-18 | 无锡风繁伟业科技有限公司 | A kind of non-intrusive electrical load alert detecting method and system |
| CN116205512A (en) * | 2022-12-16 | 2023-06-02 | 国网北京市电力公司 | Power quality warning method and device and nonvolatile storage medium |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130080103A (en) * | 2012-01-04 | 2013-07-12 | 삼성전자주식회사 | Self-diagnostic apparatus and method for electric device |
| EP2991235A1 (en) * | 2014-08-28 | 2016-03-02 | Thomson Licensing | Method of monitoring and corresponding electronic device, computer readable program product and computer readable storage medium |
| IT202200019827A1 (en) * | 2022-09-27 | 2024-03-27 | Midori S R L | PROCEDURE FOR DETECTING ANOMALIES IN ELECTRICAL EQUIPMENT, CORRESPONDING SYSTEM AND IT PRODUCT |
| CN116400126B (en) * | 2023-06-08 | 2023-08-01 | 广东佰林电气设备厂有限公司 | Low-voltage power box with data processing system |
| CN116840606B (en) * | 2023-09-01 | 2023-11-17 | 国网浙江省电力有限公司余姚市供电公司 | Power consumption abnormality monitoring method based on non-invasive load |
| CN117791872A (en) * | 2023-12-29 | 2024-03-29 | 海南英铭电力工程有限公司 | Intelligent power supply system and equipment of distribution box |
-
2024
- 2024-06-13 CN CN202410756704.5A patent/CN118330376B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109239494A (en) * | 2018-09-21 | 2019-01-18 | 无锡风繁伟业科技有限公司 | A kind of non-intrusive electrical load alert detecting method and system |
| CN116205512A (en) * | 2022-12-16 | 2023-06-02 | 国网北京市电力公司 | Power quality warning method and device and nonvolatile storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118330376A (en) | 2024-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111669375B (en) | Online safety situation assessment method and system for power industrial control terminal | |
| CN116435634B (en) | Storage battery temperature state monitoring and management system | |
| CN106934421A (en) | Converting station electric power transformer fault detecting system and detection method based on 2DPCA and SVM | |
| CN114386537B (en) | Lithium battery fault diagnosis method and device based on Catboost and electronic equipment | |
| CN118300273B (en) | Cable room fault early warning system based on multi-field coupling | |
| CN118376936B (en) | Intelligent diagnosis method and system for lithium battery state | |
| Ginart et al. | Automated feature selection for embeddable prognostic and health monitoring (PHM) architectures | |
| CN118330376B (en) | Power box with self-diagnosis abnormal electricity utilization module | |
| CN111398750A (en) | Arc identification method and system for arc identification | |
| CN111157045A (en) | Artificial intelligence electric line abnormal risk degree monitoring method and system | |
| CN116923188B (en) | Charging protection system of new energy vehicle | |
| CN116796261B (en) | Closed switch equipment mechanical characteristic prediction method based on artificial intelligence | |
| CN114781165A (en) | Multi-feature information fusion power transformer winding health diagnosis method | |
| CN115856390A (en) | Monitoring and diagnosing method and system of power switch cabinet and storage medium | |
| CN115684828A (en) | 5G leakage current fault monitoring method based on improved VMD and LMS | |
| CN111980900B (en) | Water pump fault diagnosis method based on multi-source data fusion analysis | |
| CN115219184A (en) | Acoustic fingerprint-based circuit breaker mechanical state detection system and method | |
| CN116643165B (en) | Method for detecting loss-of-magnetism fault of permanent magnet outer rotor roller motor | |
| CN118395365B (en) | Power box with function of monitoring abnormal metering data and monitoring method | |
| CN112327084B (en) | Method and system for detecting vibration and sound of running state of transformer by utilizing equidistant transformation | |
| CN118501720B (en) | Lithium battery abnormality detection system based on big data analysis | |
| CN118033297B (en) | Monitoring method of multi-mode intelligent grounding box | |
| Quek | Branch Circuit Load Monitoring using Naïve Bayesian Classifier | |
| CN118777906A (en) | Intelligent management method based on battery | |
| CN118777970A (en) | State monitoring system of electric energy metering device |
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 |