CN115456097A - Power utilization detection method and detection terminal suitable for high-power-supply low-count special transformer users - Google Patents
Power utilization detection method and detection terminal suitable for high-power-supply low-count special transformer users Download PDFInfo
- Publication number
- CN115456097A CN115456097A CN202211156755.1A CN202211156755A CN115456097A CN 115456097 A CN115456097 A CN 115456097A CN 202211156755 A CN202211156755 A CN 202211156755A CN 115456097 A CN115456097 A CN 115456097A
- Authority
- CN
- China
- Prior art keywords
- data
- detection
- abnormal
- sequence
- power
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 100
- 230000002159 abnormal effect Effects 0.000 claims abstract description 70
- 230000005611 electricity Effects 0.000 claims abstract description 58
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000010200 validation analysis Methods 0.000 claims description 3
- 238000011084 recovery Methods 0.000 claims description 2
- 230000005856 abnormality Effects 0.000 abstract description 10
- 238000004891 communication Methods 0.000 abstract description 4
- 238000005259 measurement Methods 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 description 13
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Measurement Of Current Or Voltage (AREA)
Abstract
The invention discloses a power consumption detection method and a detection terminal suitable for high-power supply and low-count special transformer users, which comprises the steps of preprocessing input data and outputting detection data; inputting the detection data into a power utilization unbalance characteristic model, a voltage abnormality characteristic model and a current abnormality characteristic model to obtain a power utilization unbalance characteristic sequence, a voltage abnormality characteristic sequence and a current abnormality characteristic sequence; obtaining an abnormal index sequence; determining an anomaly threshold p if t n If the power consumption is more than p, judging that the power consumption is abnormal; the invention solves the problems of inaccurate or missing electrical measurement data caused by unstable data communication, failure of an electrical acquisition system, system storage and the like by preprocessing data, detects electricity utilization imbalance, voltage abnormality and current abnormality by an electricity utilization detection model, obtains an abnormal index sequence, and compares an abnormal index with an abnormal threshold value to judge whether electricity utilization abnormality exists or not。
Description
Technical Field
The invention relates to the field of electric power, in particular to an electricity utilization detection method and a detection terminal suitable for high-power-supply and low-count special transformer users.
Background
The loss of the power grid can be divided into technical loss and non-technical loss. Technical losses result from heat dissipation during power transmission. And the non-technical loss is mainly caused by illegal electricity utilization behavior of users, namely power theft. The illegal electricity utilization behavior not only harms the safe and stable operation of the power grid, but also seriously harms the economic benefit of the power company. How to effectively detect illegal electricity utilization behaviors, reduce economic losses caused by illegal electricity utilization and maintain electricity utilization order is always the key point of research of scholars at home and abroad.
Traditionally, illegal persons can steal electricity by means of tampering with instruments, private wire pulling and the like. In recent years, with the construction of a smart grid, an Advanced Metering Infrastructure (AMI) in cooperation with a smart meter can automatically report power consumption information of a user at regular intervals, so that the purposes of supervision and charging are achieved.
At present, a user power utilization acquisition system can analyze and record abnormal power utilization to a certain degree, related departments are reported through interfaces, a certain number of abnormal power utilization users can be explored through technical analysis of the data, the power utilization situation of key users is tracked, but with diversification of abnormal power utilization modes under an AMI system, power stealing behaviors are more and more concealed, the degree of distinction between the abnormal power utilization behaviors and normal power utilization behaviors is less and more reduced, and the difficulty in management is increased.
Disclosure of Invention
The invention aims to solve the technical problem that the detection difficulty of abnormal electricity utilization behaviors is higher, and aims to provide an electricity utilization detection method and a detection terminal which are suitable for high-supply low-count special transformer users, so that the abnormal electricity utilization behaviors and the normal electricity utilization behaviors are effectively distinguished.
The invention is realized by the following technical scheme:
a power utilization detection method suitable for high-power supply low-count special transformer users comprises the following steps:
preprocessing input data through a data preprocessing model and outputting detection data;
inputting the detection data into a power consumption detection model, and outputting abnormal data after detection;
the detection method of the electricity utilization detection model comprises the following steps:
inputting the detection data into the power utilization unbalance characteristic model to obtain a power utilization unbalance characteristic sequence;
inputting the detection data into a voltage abnormal characteristic model to obtain a voltage abnormal characteristic sequence;
inputting the detection data into a current abnormal characteristic model to obtain a current abnormal characteristic sequence;
obtaining an abnormal index sequence through the piezoelectric unbalance characteristic sequence, the voltage abnormal characteristic sequence and the current abnormal characteristic sequence;
determining an anomaly threshold p if t n If the power consumption is more than p, the power consumption is judged to be abnormal, wherein t n Are elements in the sequence of abnormal indices.
Specifically, the method for constructing the data preprocessing model comprises the following steps:
recovery of lost data in input data:
wherein z is j Electrical parameter data representing a sampling instant, naN representing an undefined or missing value of the data;
correction of outliers in input data:
wherein mean (-) is an average value, std (-) is a standard deviation, and z is electrical reference data of each sampling moment;
normalizing the electrical reference data and outputting detection data:
specifically, the method for constructing the power utilization unbalance feature model comprises the following steps:
establishing a voltage unbalance equation and a current unbalance equation:
wherein U is in Representing the voltage value of the I-phase at time n, I in Denotes the current value, x, of the i-phase at time n Vn Representing the degree of voltage imbalance, x, at time n In Representing the current unbalance degree at the time n, wherein A, B and C represent three phases of alternating current;
build up uneven electricityConstant characteristic matrix X n ：
Wherein m is the total sampling time;
acquiring two-dimensional observation points corresponding to each row of the electricity utilization unbalance feature matrix, and establishing a sample space;
presetting a Kth distance, wherein the Kth distance is the distance between each observation point and the K-th nearest observation point, performing outlier analysis on the observation points through a local outlier factor detection algorithm, and obtaining a local abnormal factor LOF corresponding to the detection data of the time n n ；
Traversing all detection data from time 1 to time m to obtain a power utilization unbalance characteristic sequence X = (X) 1 ,x 2 ,…,x m ) In which LOF n ＝x n 。
Specifically, the method for constructing the voltage anomaly characteristic model comprises the following steps:
a1, judging whether the voltage is lost at the time n, and if the voltage is not lost, determining the deviation value of the rated voltage and the voltage in the detection data
A2, if the voltage is lost at the moment n, judging whether the moment n +1 is lost, whether the moment n +2 is lost, 8230, and whether the moment n + B is lost, wherein B is the number of moments in the set observation time B;
a3, if the judgment in the step A2 is yes, enabling y n =1; otherwise let y n ＝0
A4, traversing all the detection data from the moment 1 to the moment m to obtain a voltage abnormity characteristic sequence Y = (Y) 1 ,y 2 ,…,y m )。
Specifically, the method for constructing the current anomaly characteristic model comprises the following steps:
acquiring a clustering center curve of each phase current in the historical record, and acquiring a current record value I of each phase current at the moment n An·z 、I Bn·z 、I Cn·z ；
Obtaining a time nCurrent value of each phase, I An 、I Bn 、I Cn ；
Determining a current offset value z n ＝max(|I An·z -I An |,|I Bn·z -I Bn |,|I Cn·z -I Cn |)；
Traversing all detection data from time 1 to time m to obtain a current abnormal characteristic sequence Z = (Z) 1 ,z 2 ,…,z m )。
Specifically, the method for obtaining the abnormal index sequence comprises the following steps:
normalizing the sequence X and the sequence Z to obtain a sequence
And sequence
Obtaining an abnormal index sequence T = (T) 1 ,t 2 ,…,t m ) In which
Specifically, the training method of the electricity utilization detection model comprises the following steps:
setting K, B, B, p, h (n+1)-n A value of (a), wherein h (n+1)-n The time interval between time n and time n + 1;
inputting a training sample into the power utilization detection model, wherein the training sample comprises power utilization normal data and power utilization abnormal data;
after training is finished, inputting the verification sample into the electricity utilization detection model, and constructing a confusion matrix of a verification result;
obtaining the F1 value of the confusion matrix and comparing the value with the set F1 0 Value comparison, if F1 < F1 0 Then, reset the settings K, B, B, p, h (n+1)-n Taking values and carrying out model training again; if F1 is not less than F1 0 And outputting the electricity utilization training model.
Preferably, K =5,b =4 hours, h (n+1)-n In a time period of =15 minutes,F1 0 =0.82, number of training samples: validation sample number =7:3.
a power detection terminal suitable for users with high or low specific power supply capacity comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps of the power detection method suitable for the users with high or low specific power supply capacity.
A computer-readable storage medium, storing a computer program which, when executed by a processor, performs the steps of the power usage detection method as described above, which is suitable for high-supply low-count specific transformer users.
Compared with the prior art, the invention has the following advantages and beneficial effects:
the invention solves the problems of inaccurate or missing electrical measurement data caused by unstable data communication, faults of an electrical acquisition system, system storage and the like by preprocessing data, detects electricity utilization unbalance, voltage abnormity and current abnormity by an electricity utilization detection model, obtains an abnormal index sequence, and judges whether electricity utilization abnormity exists or not by comparing the abnormal index with an abnormal threshold value.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the principles of the invention.
Fig. 1 is a schematic flow chart of power utilization detection suitable for users with high power supply and low power consumption special transformers according to the invention.
FIG. 2 is a flow chart of the training of the power usage detection model according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant disclosure and are not to be considered as limiting.
It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings.
In the present invention, the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
Example one
As shown in fig. 1, the present embodiment provides a power consumption detection method suitable for users with high power supply and low power consumption, including:
preprocessing input data through a data preprocessing model, and outputting detection data; reasons for the need to preprocess the data: because of the problems of unstable data communication, failure of an electric acquisition system and the like, storage of the system and the like, the problems of inaccuracy and loss of electric measurement data often occur. Therefore, the data needs to be processed through a data preprocessing model, lost data is recovered, and inaccurate values are corrected.
Inputting the detection data into a power consumption detection model, and outputting abnormal data after detection; the detection data is preprocessed electricity acquisition data of a special user with high or low power supply in a period of time. By inputting the relevant data into the electricity utilization detection model, the electricity utilization abnormity can be judged through the electricity utilization detection model.
The detection method of the electricity utilization detection model comprises the following steps:
inputting the detection data into the power utilization unbalance characteristic model to obtain a power utilization unbalance characteristic sequence; when the electricity utilization is normal, the electricity utilization on the three phases is balanced, and only slight fluctuation can exist, but if the electricity utilization is abnormal, the fluctuation can be large, so that the data of the three phases are judged through the characteristic model of the electricity utilization unbalance.
Inputting the detection data into a voltage abnormal characteristic model to obtain a voltage abnormal characteristic sequence; when the voltage is abnormally reduced, electricity stealing behavior can occur, so that voltage abnormity detection is carried out on the voltage abnormity characteristic model.
Inputting the detection data into a current abnormal characteristic model to obtain a current abnormal characteristic sequence; under the condition that the power and the voltage are relatively constant, the purpose of reducing the electricity charge metering can be achieved by changing the current, and the current is subjected to abnormal detection through a current abnormal model.
Obtaining an abnormal index sequence through the piezoelectric unbalance characteristic sequence, the voltage abnormal characteristic sequence and the current abnormal characteristic sequence;
determining an anomaly threshold p if t n If the power consumption is more than p, the power consumption is judged to be abnormal, wherein t n Are elements in the sequence of abnormal indices. The sequence of abnormality indices is ordered according to a time sequence and, in practice, the electrical acquisition data is sampled at intervals over a period of time, so t n If the abnormal index is larger than the abnormal threshold value, the fact that the electricity consumption of the user is abnormal at the moment is proved, and then relevant data are output; and if the abnormality index is smaller than the abnormality threshold value, the electricity utilization of the user at the moment is proved to be normal.
Example two
This embodiment explains a method for constructing a data preprocessing model in the first embodiment.
And recovering the lost data by using an interpolation method and a filling method, and recovering the lost data in the input data:
wherein z is j Electrical parameter data representing a sampling instant, naN representing an undefined or missing value of the data; if z is j Is any character other than a number, and is denoted by NaN.
Furthermore, there may be cases where the data is inaccurate for other reasons, so it is necessary to correct outliers in the input data:
wherein mean (-) is an average value, std (-) is a standard deviation, and z is electrical reference data of each sampling moment;
to electrical reference numberData normalization is the basic process of data mining and is particularly effective for distance-based classification. Because the neural network is sensitive to features of different dimensions, in order to accelerate convergence speed and eliminate the influence of unit and scale differences among the features, normalization processing needs to be performed on the features, and detection data are output:
EXAMPLE III
This embodiment is an explanation of the method for constructing the electricity consumption detection model in the first embodiment.
The method for constructing the power utilization unbalance feature model comprises the following steps:
the electricity stealing behavior of the special transformer user of the three-phase electricity utilization can be reflected from the deviation of the imbalance rate of the voltage and the current
Establishing a voltage unbalance equation and a current unbalance equation:
wherein U is in Indicating the voltage value of the I-phase at time n, I in Denotes the current value, x, of the i-phase at time n Vn Representing the voltage unbalance, x, at time n In Representing the current unbalance degree at the time n, wherein A, B and C represent three phases of alternating current;
establishing a power utilization unbalance feature matrix X n ：
Wherein m is the total sampling time;
acquiring two-dimensional observation points corresponding to each row of the electricity utilization unbalance feature matrix, and establishing a sample space;
presetting a Kth distance, wherein the Kth distance is the distance between each observation point and the K-th nearest observation point, and then performing outlier analysis on the plurality of observation points through a local outlier factor detection algorithm, wherein the outlier refers to the observation pointThe local anomaly factor LOF corresponding to the detection data at time n is obtained from data significantly inconsistent with other data characteristics in the sample space n ；
If the electricity consumption behavior of the user is normal, the two-dimensional observation points corresponding to each line of the electricity consumption imbalance characteristic matrix are densely clustered in the sample space, the observation points which are obviously deviated from the clusters do not exist, and the local outlier factor, namely LOF (loss of tolerance) at the moment n A value close to 1; when abnormal electricity utilization occurs, the unbalance degree of the voltage and the current generates large deviation, LOF n The values are very large, thereby forming outliers.
Traversing all detection data from time 1 to time m to obtain a power utilization unbalance characteristic sequence X = (X) 1 ,x 2 ,…,x m ) In which LOF n ＝x n 。
The method for constructing the voltage anomaly characteristic model comprises the following steps:
the voltage abnormity is represented by that the voltage recorded value deviates from a rated value or tends to zero at a certain time, and the abnormal reduction of the voltage value can generally judge that the electricity stealing behavior occurs.
A1, judging whether the voltage is lost at the time n, and if the voltage is not lost, determining the deviation value of the rated voltage and the voltage in the detection data
A2, if the voltage is lost at the moment n, judging whether the voltage is lost at the moment n +1 or not, whether the voltage is lost at the moment n +2 or not, \ 8230, and whether the voltage is lost at the moment n + B or not, wherein B is the moment number in the set observation time B;
the voltage loss record of the user electric energy metering device can have two conditions, namely disconnection of a metering loop or power supply abnormity, and the two conditions can occur that the user is normally used but the voltage loss occurs. Therefore, in order to eliminate the above-described situation, one observation time B is set, and it is determined that the pressure is lost at a plurality of times within the observation time, and it is determined that the pressure is abnormal.
A3, if the judgment in the step A2 is yes, making y n =1; otherwise let y n ＝0
A4, traversing time 1Obtaining a voltage abnormity characteristic sequence Y = (Y) by all detection data in time m 1 ,y 2 ,…,y m )。
The method for constructing the current abnormal feature model comprises the following steps:
for a special transformer user, the daily power load curves are similar, so that the daily current curves of normal power users are determined to have similarity, a clustering center curve of each phase current in a historical record is obtained, and a current record value I of each phase current at the moment n is obtained An·z 、I Bn·z 、I Cn·z ；
In the embodiment, the records of the previous 20 days can be obtained, the clustering center curve can be obtained, and the current record value can be obtained according to the coordinates.
Obtaining the current value of each phase in time n, I An 、I Bn 、I Cn ；
Determining a current offset value z n ＝max(|I An·z -I An |,|I Bn·z -I Bn |,|I Cn·z -I Cn |)；
Traversing all detection data from time 1 to time m to obtain a current abnormal characteristic sequence Z = (Z) 1 ,z 2 ,…,z m )。
The method for obtaining the abnormal index sequence comprises the following steps:
normalizing the sequence X and the sequence Z to eliminate the influence of the dimension on the calculation result and obtain the sequence
And sequence
Obtaining the abnormal index sequence T = (T) 1 ,t 2 ,…,t m ) In which
After the power consumption detection model is constructed, the set value needs to be trained, as shown in fig. 2, the method for training the power consumption detection model in this embodiment includes:
setting K, B, B, p, h (n+1)-n A value of (a), wherein h (n+1)-n The time interval between time n and time n + 1;
inputting a training sample into the electricity utilization detection model, wherein the training sample comprises electricity utilization normal data and electricity utilization abnormal data;
after training is finished, inputting the verification sample into the electricity utilization detection model, and constructing a confusion matrix of a verification result;
the essence of the abnormal electricity utilization detection process is a binary classification problem, all samples are classified into a positive type (abnormal data) or a negative type (normal data), and when the positive type samples and the negative type samples in the detected samples are extremely unbalanced in distribution, the detection effect of the samples is meaningless by directly using the accuracy of the detection result.
Obtaining the F1 value of the confusion matrix and comparing the value with the set F1 0 Value comparison, if F1 < F1 0 Then, the settings K, B, B, p, h are reset (n+1)-n Taking values and carrying out model training again; if F1 is not less than F1 0 And outputting the electricity utilization training model.
To provide a preferred set point, K =5, b =4 hours, h (n+1)-n =15 min, F1 0 =0.82, number of training samples: validation sample number =7:3.
EXAMPLE III
The power utilization detection terminal suitable for the high-power supply low-count specific power transformer user comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, and the steps of the power utilization detection method suitable for the high-power supply low-count specific power transformer user are realized when the processor executes the computer program.
The memory may be used to store software programs and modules, and the processor may execute various functional applications of the terminal and data processing by operating the software programs and modules stored in the memory. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an execution program required for at least one function, and the like.
The storage data area may store data created according to the use of the terminal, and the like. Further, the memory may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
A computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the power usage detection method as described above, which is suitable for users with high power supply and low power consumption.
Without loss of generality, computer readable media may comprise computer storage media and communication media. Computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instruction data structures, program modules or other data. Computer storage media includes RAM, ROM, EPROM, EEPROM, flash memory or other solid state memory technology, CD-ROM, DVD, or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art will appreciate that computer storage media is not limited to the foregoing. The system memory and mass storage devices described above may be collectively referred to as memory.
In the description herein, reference to the description of the terms "one embodiment/mode," "some embodiments/modes," "example," "specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment/mode or example is included in at least one embodiment/mode or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to be the same embodiment/mode or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments/modes or examples. Furthermore, the various embodiments/aspects or examples and features of the various embodiments/aspects or examples described in this specification can be combined and combined by one skilled in the art without conflicting therewith.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or to implicitly indicate the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
It will be appreciated by those skilled in the art that the above embodiments are only for clarity of illustration of the invention, and are not intended to limit the scope of the invention. It will be apparent to those skilled in the art that other variations or modifications may be made on the above invention and still be within the scope of the invention.
Claims (10)
1. A power utilization detection method suitable for high-power supply low-count special transformer users is characterized by comprising the following steps:
preprocessing input data through a data preprocessing model and outputting detection data;
inputting the detection data into a power consumption detection model, and outputting abnormal data after detection;
the detection method of the electricity utilization detection model comprises the following steps:
inputting the detection data into the power utilization unbalance characteristic model to obtain a power utilization unbalance characteristic sequence;
inputting the detection data into a voltage abnormal characteristic model to obtain a voltage abnormal characteristic sequence;
inputting the detection data into a current abnormal characteristic model to obtain a current abnormal characteristic sequence;
obtaining an abnormal index sequence through the piezoelectric unbalance characteristic sequence, the voltage abnormal characteristic sequence and the current abnormal characteristic sequence;
determining an anomaly threshold p if t n If the power consumption is more than p, the power consumption is judged to be abnormal, wherein t n Are elements in the sequence of abnormal indices.
2. The power consumption detection method suitable for the high-supply low-count special transformer users as claimed in claim 1, wherein the method for constructing the data preprocessing model comprises the following steps:
recovery of lost data in input data:
wherein z is j Electrical parameter data representing a sampling instant, naN representing an undefined or missing value of the data;
correcting outliers in the input data:
wherein mean (-) is an average value, std (-) is a standard deviation, and z is electrical reference data of each sampling moment;
normalizing the electrical reference data and outputting detection data:
3. the method for detecting the electricity consumption of the user with the high supply and low metering special transformer as claimed in claim 1, wherein the method for constructing the electricity consumption imbalance characteristic model comprises the following steps:
establishing a voltage unbalance equation and a current unbalance equation:
wherein U is in Indicating the voltage value of the I-phase at time n, I in Denotes the current value, x, of the i-phase at time n Vn Representing the degree of voltage imbalance, x, at time n In Representing the current unbalance degree at the time n, wherein A, B and C represent three phases of alternating current;
establishing a power utilization unbalance feature matrix X n ：
Wherein m is the total sampling time;
acquiring two-dimensional observation points corresponding to each row of the electricity utilization unbalance feature matrix, and establishing a sample space;
presetting a Kth distance, wherein the Kth distance is the distance between each observation point and the K-th nearest observation point, performing outlier analysis on the observation points through a local outlier factor detection algorithm, and obtaining local abnormal factors LOF corresponding to the detection data of the moment n n ；
Traversing all detection data from time 1 to time m to obtain a power utilization unbalance characteristic sequence X = (X) 1 ,x 2 ,…,x m ) In which LOF n ＝x n 。
4. The method for detecting the electricity consumption of the user with the high supply and low metering transformer as claimed in claim 3, wherein the method for constructing the voltage anomaly characteristic model comprises the following steps:
a1, judging whether the voltage is lost at the time n, and if the voltage is not lost, determining the deviation value of the rated voltage and the voltage in the detection data
A2, if the voltage is lost at the moment n, judging whether the moment n +1 is lost, whether the moment n +2 is lost, 8230, and whether the moment n + B is lost, wherein B is the number of moments in the set observation time B;
a3, if the judgment in the step A2 is yes, making y n =1; otherwise let y n ＝0
And A4, traversing all the detection data from the moment 1 to the moment m to obtain a voltage abnormity feature sequence Y = (Y) 1 ,y 2 ,…,y m )。
5. The power utilization detection method suitable for the high-supply low-count special transformer users as claimed in claim 4, wherein the method for constructing the current anomaly characteristic model comprises the following steps:
acquiring a clustering center curve of each phase current in the historical record, and acquiring a current record value I of each phase current at the moment n An·z 、I Bn·z 、I Cn·z ；
Obtaining the current value, I, of each phase at time n An 、I Bn 、I Cn ；
Determining a current offset value z n ＝max(|I An·z -I An |,|I Bn·z -I Bn |,|I Cn·z -I Cn |)；
Traversing all detection data from time 1 to time m to obtain a current abnormal characteristic sequence Z = (Z) 1 ,z 2 ,…,z m )。
6. The power consumption detection method suitable for the high-supply low-count special transformer users as claimed in claim 5, wherein the method for obtaining the abnormal index sequence comprises the following steps:
normalizing the sequence X and the sequence Z to obtain a sequence
And sequence
Obtaining the abnormal index sequence T = (T) 1 ,t 2 ,…,t m ) Wherein
7. The power consumption detection method suitable for the high-power supply low-count special transformer users as claimed in claim 6, wherein the training method of the power consumption detection model comprises the following steps:
setting K, B, B, p, h (n+1)-n A value of (a), wherein h (n+1)-n The time interval between time n and time n + 1;
inputting a training sample into the power utilization detection model, wherein the training sample comprises power utilization normal data and power utilization abnormal data;
after training is finished, inputting the verification sample into the electricity utilization detection model, and constructing a confusion matrix of a verification result;
obtaining the F1 value of the confusion matrix and comparing the value with the set F1 0 Value comparison, if F1 < F1 0 Then, reset the settings K, B, B, p, h (n+1)-n Taking values and carrying out model training again; if F1 is not less than F1 0 And outputting the electricity utilization training model.
8. The method as claimed in claim 7, wherein K =5,B =4 hours, h (n+1)-n =15 min, F1 0 =0.82, number of training samples: validation sample number =7:3.
9. an electricity detection terminal suitable for users with high power supply and low power consumption, comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to realize the steps of the electricity detection method suitable for users with high power supply and low power consumption according to any one of claims 1 to 8.
10. A computer-readable storage medium, in which a computer program is stored, wherein the computer program, when being executed by a processor, implements the steps of the power detection method for users with high or low design variables according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211156755.1A CN115456097A (en) | 2022-09-22 | 2022-09-22 | Power utilization detection method and detection terminal suitable for high-power-supply low-count special transformer users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211156755.1A CN115456097A (en) | 2022-09-22 | 2022-09-22 | Power utilization detection method and detection terminal suitable for high-power-supply low-count special transformer users |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115456097A true CN115456097A (en) | 2022-12-09 |
Family
ID=84307517
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211156755.1A Pending CN115456097A (en) | 2022-09-22 | 2022-09-22 | Power utilization detection method and detection terminal suitable for high-power-supply low-count special transformer users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115456097A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117491792A (en) * | 2023-12-27 | 2024-02-02 | 四川中威能电力科技有限公司 | Power consumption abnormality detection method, system and storage medium based on intelligent ammeter |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102570889A (en) * | 2011-01-04 | 2012-07-11 | 西华大学 | Three-phase four-bridge arm inverter and control method |
| CN103760462A (en) * | 2014-01-14 | 2014-04-30 | 昆明理工大学 | Current transient quantity principal component cluster analysis direction protection method for power transmission line with static synchronous series compensator |
| CN204086389U (en) * | 2014-08-28 | 2015-01-07 | 国家电网公司 | A kind of three-phase electricity monitoring device |
| CN104407268A (en) * | 2014-11-27 | 2015-03-11 | 国家电网公司 | Abnormal electricity utilization judgment method based on abnormal analysis of electric quantity, voltage and current |
| WO2019237492A1 (en) * | 2018-06-13 | 2019-12-19 | 山东科技大学 | Semi-supervised learning-based abnormal electricity utilization user detection method |
| CN111932078A (en) * | 2020-07-10 | 2020-11-13 | 国网浙江省电力有限公司绍兴供电公司 | Risk user identification method based on measurement data multi-situation evaluation |
| CN112766342A (en) * | 2021-01-12 | 2021-05-07 | 安徽容知日新科技股份有限公司 | Abnormity detection method for electrical equipment |
| CN113030618A (en) * | 2021-03-15 | 2021-06-25 | 广东电网有限责任公司 | Island detection method and device of power generation system and power generation system |
-
2022
- 2022-09-22 CN CN202211156755.1A patent/CN115456097A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102570889A (en) * | 2011-01-04 | 2012-07-11 | 西华大学 | Three-phase four-bridge arm inverter and control method |
| CN103760462A (en) * | 2014-01-14 | 2014-04-30 | 昆明理工大学 | Current transient quantity principal component cluster analysis direction protection method for power transmission line with static synchronous series compensator |
| CN204086389U (en) * | 2014-08-28 | 2015-01-07 | 国家电网公司 | A kind of three-phase electricity monitoring device |
| CN104407268A (en) * | 2014-11-27 | 2015-03-11 | 国家电网公司 | Abnormal electricity utilization judgment method based on abnormal analysis of electric quantity, voltage and current |
| WO2019237492A1 (en) * | 2018-06-13 | 2019-12-19 | 山东科技大学 | Semi-supervised learning-based abnormal electricity utilization user detection method |
| CN111932078A (en) * | 2020-07-10 | 2020-11-13 | 国网浙江省电力有限公司绍兴供电公司 | Risk user identification method based on measurement data multi-situation evaluation |
| CN112766342A (en) * | 2021-01-12 | 2021-05-07 | 安徽容知日新科技股份有限公司 | Abnormity detection method for electrical equipment |
| CN113030618A (en) * | 2021-03-15 | 2021-06-25 | 广东电网有限责任公司 | Island detection method and device of power generation system and power generation system |
Non-Patent Citations (1)
| Title |
|---|
| 郭思涵: ""基于用电特征的窃电行为挖掘策略研究"", 《CNKI优秀硕士学位论文全文库 工程科技Ⅱ辑》, no. 03, pages 1 - 72 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117491792A (en) * | 2023-12-27 | 2024-02-02 | 四川中威能电力科技有限公司 | Power consumption abnormality detection method, system and storage medium based on intelligent ammeter |
| CN117491792B (en) * | 2023-12-27 | 2024-03-22 | 四川中威能电力科技有限公司 | Power consumption abnormality detection method, system and storage medium based on intelligent ammeter |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112699913B (en) | Method and device for diagnosing abnormal relationship of household transformer in transformer area | |
| CN105825298B (en) | Power grid metering early warning system and method based on load characteristic estimation | |
| CN108490288B (en) | A kind of stealing detection method and system | |
| CN111985824A (en) | Non-invasive load monitoring method and monitoring equipment for intelligent ammeter box | |
| CN114386537B (en) | Lithium battery fault diagnosis method and device based on Catboost and electronic equipment | |
| US20200217896A1 (en) | Device and method for evaluating energy storage device and evaluation system | |
| CN112198375B (en) | Phase identification method, device, equipment and storage medium for single-phase user | |
| CN115456097A (en) | Power utilization detection method and detection terminal suitable for high-power-supply low-count special transformer users | |
| CN108764603A (en) | A kind of user's electricity stealing discrimination method based on higher-dimension random matrix | |
| CN110967585B (en) | Malignant load identification method and device | |
| CN116775731A (en) | Sliding window-based calculation method for fault-connection trace-back electric quantity of three-phase electric energy meter | |
| CN109142830A (en) | Stealing detection method based on power information acquisition system big data | |
| CN117368745B (en) | Hard-pack lithium battery safety monitoring method and device based on deep learning | |
| Shuai et al. | Intelligent diagnosis of abnormal charging for electric bicycles based on improved dynamic time warping | |
| CN111366889B (en) | Abnormal electricity utilization detection method for intelligent electric meter | |
| CN117169761A (en) | Battery state evaluation method, apparatus, device, storage medium, and program product | |
| CN117331017A (en) | Method and system for studying and judging misconnection of three-phase four-wire electric energy meter | |
| CN111476316B (en) | Method and system for clustering mean value of power load characteristic data based on cloud computing | |
| CN112948742A (en) | Non-household load identification method based on average load and total demand distortion rate | |
| CN118152836B (en) | Stability evaluation method for operation process of electric energy meter | |
| CN112731261B (en) | Fault detection method and device for electric energy metering equipment, computer equipment and medium | |
| CN117541097A (en) | Digital power data analysis method and system | |
| Zhang et al. | A Novel Method of Phase Identification in Low Voltage Distribution Network | |
| Sun et al. | Research on Typical Load Classification of Electricity Users Based on Bad Data Detection | |
| López et al. | A comparison of placement methods for collecting PMU data used in angular stability detection |
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 |