CN112819373A - Distribution network voltage abnormal data detection method and device - Google Patents
Distribution network voltage abnormal data detection method and device Download PDFInfo
- Publication number
- CN112819373A CN112819373A CN202110214662.9A CN202110214662A CN112819373A CN 112819373 A CN112819373 A CN 112819373A CN 202110214662 A CN202110214662 A CN 202110214662A CN 112819373 A CN112819373 A CN 112819373A
- Authority
- CN
- China
- Prior art keywords
- voltage
- value
- distribution network
- actual
- values
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 72
- 238000001514 detection method Methods 0.000 title claims abstract description 39
- 238000005259 measurement Methods 0.000 claims abstract description 108
- 238000000034 method Methods 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims description 22
- 238000013528 artificial neural network Methods 0.000 claims description 14
- 238000012360 testing method Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 abstract description 3
- 238000003062 neural network model Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 4
- 238000005065 mining Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
- G06Q10/063114—Status monitoring or status determination for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Measurement Of Current Or Voltage (AREA)
Abstract
The application relates to the technical field of big data analysis of power systems, and discloses a distribution network voltage abnormal data detection method and device. In the method, a Bi-LSTM neural network model is trained by using historical distribution network voltage data, and a distribution network voltage prediction model containing preset rules is established. And then inputting a voltage data set to be detected containing a plurality of voltage actual measurement values into a distribution network voltage prediction model to obtain a voltage prediction value which corresponds to the voltage actual measurement values and accords with a preset rule, and determining a voltage error value according to the corresponding voltage actual measurement values and the voltage prediction value. And further determining a standard difference value of the voltage errors, and determining whether the actual measured value of the target voltage is abnormal data according to the correlation between the target voltage error value and the standard difference value of the voltage errors. The method and the device can effectively detect abnormal data in the power distribution network data acquisition system, and provide powerful technical data support for various aspects of the power distribution network operation process.
Description
Technical Field
The application relates to the technical field of big data analysis of power systems, in particular to a distribution network voltage abnormal data detection method and device.
Background
The distribution network voltage data is monitored and stored by a distribution network data acquisition system. The distribution network voltage data acquisition system uploads the distribution network voltage data once every a period of time, so a large amount of historical distribution network voltage data can be accumulated in a database of the distribution network voltage acquisition system. The voltage data of the distribution network comprises the actual voltage measurement value, and if the actual voltage measurement value can be deeply mined and researched, powerful technical data support can be brought to the aspects of power utilization prediction, abnormal detection, safe operation, decision scheduling and the like in the operation process of the distribution network.
The currently stored distribution network voltage data has the characteristics of multiple sources, heterogeneous information, large quantity, various attributes and the like, and often has more abnormal data. Therefore, before information mining is performed by using the distribution network voltage data, abnormal data in the distribution network voltage data needs to be detected, so that operators can obtain correct and reliable distribution network voltage data.
Disclosure of Invention
The application discloses a distribution network voltage abnormal data detection method and device, which are used for solving the technical problems that in the prior art, more abnormal data exist in stored distribution network voltage data in a database of a distribution network voltage acquisition system, and how to detect the abnormal data is solved.
The application discloses in a first aspect a distribution network voltage abnormal data detection method, which comprises the following steps:
acquiring historical distribution network voltage data and generating a historical voltage data set;
acquiring a voltage data set to be detected according to the historical voltage data set, wherein the voltage data set to be detected comprises a plurality of voltage actual measurement values which are sequentially arranged according to a measurement time sequence;
inputting the voltage data set to be detected into a pre-constructed distribution network voltage prediction model, and obtaining a plurality of voltage prediction values output by the distribution network voltage prediction model, wherein the voltage prediction values are respectively in one-to-one correspondence with voltage actual measurement values in the voltage data set to be detected except for the first two voltage actual measurement values, and the distribution network voltage prediction model comprises a preset rule;
obtaining a plurality of voltage error values, wherein any voltage error value is a difference value between any predicted voltage value and a corresponding actual voltage measurement value;
determining a voltage error standard deviation value according to the voltage error values;
and determining whether the actual measured value of the target voltage is abnormal data or not according to the target voltage error value and the standard difference value of the voltage error, wherein the target voltage error value corresponds to the actual measured value of the target voltage, and the target voltage error value is any one of the voltage error values.
Optionally, the preset rule includes a corresponding input rule and an output rule:
the input rule is that the actual voltage measurement value at the t-2 moment and the actual voltage measurement value at the t-1 moment in the voltage data set to be detected are used as the input of the distribution network voltage prediction model;
the output rule is that the actual voltage measurement value at the t-th moment in the voltage data set to be detected is used as the output of the distribution network voltage prediction model, and the t-th moment is the moment corresponding to any actual voltage measurement value except the corresponding moments of the two previous actual voltage measurement values in the voltage data set to be detected.
Optionally, the step of constructing the distribution network voltage prediction model includes:
dividing the historical voltage data set into a training set and a test set according to a preset proportion, wherein the test set is the voltage data set to be detected;
and training the Bi-LSTM neural network prediction model by using the training set to construct a distribution network voltage prediction model.
Optionally, the obtaining a plurality of voltage error values includes:
determining the plurality of voltage error values by:
st＝|yt-xt|；
wherein s istIndicating the value of the voltage error at time t, ytIndicates the predicted value of voltage, x, at time ttIndicating the actual voltage measurement at time t.
Optionally, the determining a standard deviation value of voltage errors according to the plurality of voltage error values includes:
determining the standard deviation value of the voltage error by the following formula:
wherein μ represents an average value of the voltage error values, m represents the number of the voltage error values, t represents a time, stRepresents the voltage error value at time t, and σ represents the standard deviation value of the voltage error.
The second aspect of the present application discloses a distribution network voltage abnormal data detection device, which is applied to the distribution network voltage abnormal data detection method disclosed in the first aspect of the present application, and the distribution network voltage abnormal data detection device includes:
the historical voltage data acquisition module is used for acquiring historical distribution network voltage data and generating a historical voltage data set;
the voltage data acquisition module is used for acquiring a voltage data set to be detected according to the historical voltage data set, wherein the voltage data set to be detected comprises a plurality of voltage actual measurement values which are sequentially arranged according to a measurement time sequence;
the voltage prediction value acquisition module is used for inputting the voltage data set to be detected into a pre-constructed distribution network voltage prediction model and acquiring a plurality of voltage prediction values output by the distribution network voltage prediction model, wherein the plurality of voltage prediction values are respectively in one-to-one correspondence with voltage actual measurement values in the voltage data set to be detected except for the first two voltage actual measurement values, and the distribution network voltage prediction model comprises a preset rule;
the voltage error value acquisition module is used for acquiring a plurality of voltage error values, wherein any voltage error value is a difference value between any predicted voltage value and a corresponding actual voltage measurement value;
the voltage error standard difference value acquisition module is used for determining a voltage error standard difference value according to the plurality of voltage error values;
and the abnormal data judgment module is used for determining whether the actual measured value of the target voltage is abnormal data according to a target voltage error value and the standard voltage error difference value, wherein the target voltage error value corresponds to the actual measured value of the target voltage, and the target voltage error value is any one of the voltage error values.
Optionally, the preset rule includes a corresponding input rule and an output rule:
the input rule is that the actual voltage measurement value at the t-2 moment and the actual voltage measurement value at the t-1 moment in the voltage data set to be detected are used as the input of the distribution network voltage prediction model;
the output rule is that the actual voltage measurement value at the t-th moment in the voltage data set to be detected is used as the output of the distribution network voltage prediction model, and the t-th moment is the moment corresponding to any actual voltage measurement value except the corresponding moments of the two previous actual voltage measurement values in the voltage data set to be detected.
Optionally, the distribution network voltage abnormal data detection apparatus further includes a model building module, where the model building module is configured to pre-build the distribution network voltage prediction model, and the model building module includes:
the dividing processing unit is used for dividing the historical voltage data set into a training set and a test set according to a preset proportion, wherein the test set is the voltage data set to be detected;
and the distribution network voltage prediction model construction unit is used for training the Bi-LSTM neural network prediction model by using the training set to construct a distribution network voltage prediction model.
Optionally, the voltage error value obtaining module is configured to determine the plurality of voltage error values according to the following formula:
st＝|yt-xt|；
wherein s istIndicating the value of the voltage error at time t, ytIndicates the predicted value of voltage, x, at time ttIndicating the actual voltage measurement at time t.
Optionally, the voltage error standard deviation value obtaining module is configured to determine the voltage error standard deviation value according to the following formula:
wherein μ represents an average value of the voltage error values, m represents the number of the voltage error values, t represents a time, stRepresents the voltage error value at time t, and σ represents the standard deviation value of the voltage error.
The application relates to the technical field of big data analysis of power systems, and discloses a distribution network voltage abnormal data detection method and device. In the method, firstly, a Bi-LSTM neural network model is trained by using historical distribution network voltage data based on a Bi-LSTM neural network, and a distribution network voltage prediction model containing preset rules is established. And then inputting a voltage data set to be detected containing a plurality of voltage actual measurement values into a distribution network voltage prediction model to obtain a voltage prediction value which corresponds to the voltage actual measurement values and accords with a preset rule, and determining a voltage error value according to the corresponding voltage actual measurement values and the voltage prediction value. And further determining a standard difference value of the voltage errors, and determining whether the actual measured value of the target voltage is abnormal data according to the correlation between the target voltage error value and the standard difference value of the voltage errors. The method and the device can effectively detect the abnormal data in the power distribution network data acquisition system, and provide powerful technical data support for power utilization prediction, abnormal detection, safe operation, decision scheduling and other aspects in the power distribution network operation process.
Drawings
In order to more clearly explain the technical solution of the present application, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious to those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic view of a work flow of a distribution network voltage abnormal data detection method disclosed in an embodiment of the present application;
fig. 2 is a schematic view of a workflow of a distribution network voltage prediction model that is pre-constructed in the distribution network voltage abnormal data detection method disclosed in the embodiment of the present application;
fig. 3 is a schematic structural diagram of a distribution network voltage abnormal data detection device disclosed in an embodiment of the present application.
Detailed Description
In order to solve the technical problem that in the prior art, more abnormal data exist in the stored distribution network voltage data in a database of a distribution network voltage acquisition system, and how to detect the abnormal data, the application discloses a distribution network voltage abnormal data detection method and a distribution network voltage abnormal data detection device through the following two embodiments.
The first embodiment of the application discloses a distribution network voltage abnormal data detection method, which is shown in a working flow diagram in fig. 1 and comprises the following steps:
step S101, obtaining historical distribution network voltage data and generating a historical voltage data set.
In practical application, in order to ensure the reliability of the constructed distribution network voltage prediction model, at least 10000 actual voltage measurement values which are sequentially arranged according to the measurement time sequence are required in the historical voltage data set.
And S102, acquiring a voltage data set to be detected according to the historical voltage data set, wherein the voltage data set to be detected comprises a plurality of voltage actual measurement values which are sequentially arranged according to a measurement time sequence.
Step S103, inputting the voltage data set to be detected into a pre-constructed distribution network voltage prediction model, and obtaining a plurality of voltage prediction values output by the distribution network voltage prediction model, wherein the plurality of voltage prediction values are respectively in one-to-one correspondence with voltage actual measurement values in the voltage data set to be detected except for the first two voltage actual measurement values, and the distribution network voltage prediction model comprises a preset rule.
Further, the preset rules include corresponding input rules and output rules:
and the input rule is that the actual voltage measurement value at the t-2 moment and the actual voltage measurement value at the t-1 moment in the voltage data set to be detected are used as the input of the distribution network voltage prediction model.
The output rule is that the actual voltage measurement value at the t-th moment in the voltage data set to be detected is used as the output of the distribution network voltage prediction model, and the t-th moment is the moment corresponding to any actual voltage measurement value except the corresponding moments of the two previous actual voltage measurement values in the voltage data set to be detected.
Specifically, n (n is more than or equal to 3, and n is a positive integer) actual voltage measurement values are input, and n-2 predicted voltage values are output through the distribution network voltage prediction model. The correspondence between the actual voltage measurement value and the predicted voltage value is shown in the following example.
Example (c):
inputting a voltage data set to be detected to the distribution network voltage prediction model, wherein the voltage data set to be detected specifically comprises an actual voltage measurement value x at the t-4 th momentt-4And the actual voltage measurement value x at the t-3 th momentt-3And the actual voltage measurement value x at the t-2 th momentt-2At time t-1Actual measurement value x of voltaget-1And the actual voltage measurement x at time ttThe t-4 th time, the t-3 th time, the t-2 th time, the t-1 th time and the t-th time are sequentially arranged according to the measuring time sequence. According to the actual voltage measurement value x at the t-4 th momentt-4And the actual voltage measurement x at time t-3t-3And outputting the predicted voltage value y at the t-2 th momentt-2(ii) a According to the actual voltage measurement value x at the t-3 momentt-3And the actual voltage measurement x at time t-2t-2And outputting the predicted voltage value y at the t-1 th momentt-1(ii) a According to the actual voltage measurement value x at the t-2 momentt-2And the actual voltage measurement x at time t-1t-1Outputting the predicted voltage value y at the t-th momentt. The actual voltage measurement value and the predicted voltage value at each time point correspond to each other.
In practical application, the voltage data sets to be detected include a plurality of voltage actual measurement values sequentially arranged according to the measurement time sequence, and the rest of the voltage data sets to be detected refer to the above example and so on.
Step S104, obtaining a plurality of voltage error values, wherein any one of the voltage error values is a difference between any one of the predicted voltage values and a corresponding actual voltage measurement value.
In some embodiments of the present application, a difference between the predicted voltage value and the actual voltage measurement value at any time except the first two times in the voltage dataset to be detected is calculated, and the obtained multiple voltage error values are recorded as an error set S.
Further, the obtaining a plurality of voltage error values includes:
determining the plurality of voltage error values by:
st＝|yt-xt|
wherein s istIndicating the value of the voltage error at time t, ytIndicates the predicted value of voltage, x, at time ttIndicating the actual voltage measurement at time t.
Step S105, determining a standard deviation value of the voltage errors according to the voltage error values.
Further, the determining a standard deviation value of voltage errors according to the plurality of voltage error values includes:
determining the standard deviation value of the voltage error by the following formula:
wherein μ represents an average value of the voltage error values, m represents the number of the voltage error values, t represents a time, stRepresents the voltage error value at time t, and σ represents the standard deviation value of the voltage error.
Step S106, determining whether the actual measurement value of the target voltage is abnormal data according to a target voltage error value and the standard voltage error difference value, wherein the target voltage error value corresponds to the actual measurement value of the target voltage, and the target voltage error value is any one of the voltage error values.
In some embodiments of the present application, the error set S is modeled as a normal distribution obeying the average value μ of the voltage error values, the voltage error standard deviation value σ, and according to the 3 σ principle of the normal distribution, the time corresponding to the target voltage error value greater than the 3 σ range is regarded as an abnormal data point, and the time corresponding to the target voltage error value within the 3 σ range is regarded as a normal data point. And further determining the target voltage actual measurement value corresponding to the abnormal data point as abnormal data, and determining the target voltage actual measurement value corresponding to the normal data point as normal data.
According to the distribution network voltage abnormal data detection method disclosed by the embodiment of the application, firstly, a Bi-LSTM neural network model is trained by using historical distribution network voltage data based on a Bi-LSTM neural network, and a distribution network voltage prediction model containing preset rules is established. And then inputting a voltage data set to be detected containing a plurality of voltage actual measurement values into a distribution network voltage prediction model to obtain a voltage prediction value which corresponds to the voltage actual measurement values and accords with a preset rule, and determining a voltage error value according to the corresponding voltage actual measurement values and the voltage prediction value. And further determining a standard difference value of the voltage errors, and determining whether the actual measured value of the target voltage is abnormal data according to the correlation between the target voltage error value and the standard difference value of the voltage errors. The method and the device can effectively detect the abnormal data in the power distribution network data acquisition system, and provide powerful technical data support for power utilization prediction, abnormal detection, safe operation, decision scheduling and other aspects in the power distribution network operation process.
In some embodiments of the present application, a group of distribution network voltage data which are all normal data is used, a partial voltage actual measurement value is preset as abnormal data, and then verification is performed by using the distribution network voltage abnormal data detection method disclosed in the first embodiment of the present application, and the Accuracy, Recall, Precision and F1 values of the distribution network voltage data are calculated. The Accuracy refers to the ratio of the number of correctly detected samples to all detected samples; the Recall rate Recall refers to the ratio of the number of the correctly detected abnormal data to the number of the actual abnormal data; the discrimination Precision refers to the ratio of the number of correctly detected abnormal data to the number of detected abnormal data; the F1 value is a compromise indicator of recall and recognition, and is a harmonic mean of recall and recognition.
Specifically, the verification is performed by the following formula:
in the formula, TP represents the number of abnormal data in the actual condition and abnormal data in the detected condition, FN represents the number of abnormal data in the actual condition and normal data in the detected condition, FP represents the number of normal data in the actual condition and abnormal data in the detected condition, and TN represents the number of normal data in the actual condition and normal data in the detected condition. The specific description is shown in table 1.
TABLE 1
The larger the Accuracy, Recall, Precision and F1 values are, the better the detection effect of the distribution network voltage abnormal data detection method disclosed by the embodiment of the application is proved.
Further, referring to the workflow diagram shown in fig. 2, the step of constructing the distribution network voltage prediction model includes:
step S201, dividing the historical voltage data set into a training set and a test set according to a preset proportion, wherein the test set is the voltage data set to be detected.
In some embodiments of the present application, the preset ratio is 8: 2, when the historical voltage data set includes 10000 actual voltage measurement values, then the training set includes 8000 actual voltage measurement values arranged in sequence according to the time sequence of measurement, the test set includes 2000 actual voltage measurement values arranged in sequence according to the time sequence of measurement.
And S202, training the Bi-LSTM neural network prediction model by using the training set, and constructing a distribution network voltage prediction model.
The Bi-LSTM neural network is formed by combining a forward LSTM neural network and a backward LSTM neural network, and belongs to further improvement on the basis of an LSTM (Long Short-Term Memory) neural network.
Specifically, the training process is as follows: measuring the actual voltage value x at the t +1 th momentt+1And the actual measured value x of the voltage at the t +2 th momentt+2As the input of the Bi-LSTM neural network prediction model, the actual measured voltage value x at the t +3 th moment is usedt+3And as the output of the Bi-LSTM neural network prediction model, the t +1 th time, the t +2 th time and the t +3 th time are the time of measurement time sequence arrangement. By analogy, the distribution network voltage prediction model is constructed after 7998 times of training is completed according to the training set with the actual voltage measurement value number of 8000. By mining the information of the forward sequence and the reverse sequence, the rule of the time sequence information can be better learned, the method is more suitable for being applied to distribution network voltage abnormal data detection with time sequence characteristics, and the method has better application prospect in the aspect of distribution network voltage abnormal data detection.
The following are embodiments of the apparatus of the present application that may be used to perform embodiments of the method of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method of the present application.
The second embodiment of the present application discloses a distribution network voltage abnormal data detection device, which is applied to the distribution network voltage abnormal data detection method disclosed in the first embodiment, referring to the structural schematic diagram shown in fig. 3, the distribution network voltage abnormal data detection device includes:
the historical voltage data acquisition module 10 is used for acquiring historical distribution network voltage data and generating a historical voltage data set;
the voltage data acquisition module 20 is configured to acquire a voltage data set to be detected according to the historical voltage data set, where the voltage data set to be detected includes a plurality of actual voltage measurement values arranged in sequence according to a measurement time sequence;
a voltage prediction value obtaining module 30, configured to input the voltage data set to be detected into a pre-constructed distribution network voltage prediction model, and obtain a plurality of voltage prediction values output by the distribution network voltage prediction model, where the plurality of voltage prediction values are respectively in one-to-one correspondence with voltage actual measurement values in the voltage data set to be detected except for the first two voltage actual measurement values, and the distribution network voltage prediction model includes a preset rule;
a voltage error value obtaining module 40, configured to obtain a plurality of voltage error values, where any one of the voltage error values is a difference between any one of the predicted voltage values and a corresponding actual voltage measurement value;
a voltage error standard deviation value obtaining module 50, configured to determine a voltage error standard deviation value according to the multiple voltage error values;
the abnormal data determining module 60 is configured to determine whether the actual measured target voltage value is abnormal data according to a target voltage error value and the standard voltage error difference value, where the target voltage error value corresponds to the actual measured target voltage value, and the target voltage error value is any one of the voltage error values.
Further, the preset rules include corresponding input rules and output rules:
and the input rule is that the actual voltage measurement value at the t-2 moment and the actual voltage measurement value at the t-1 moment in the voltage data set to be detected are used as the input of the distribution network voltage prediction model.
The output rule is that the actual voltage measurement value at the t-th moment in the voltage data set to be detected is used as the output of the distribution network voltage prediction model, and the t-th moment is the moment corresponding to any actual voltage measurement value except the corresponding moments of the two previous actual voltage measurement values in the voltage data set to be detected.
Further, the distribution network voltage abnormal data detection device further comprises a model building module, wherein the model building module is used for building the distribution network voltage prediction model in advance, and the model building module comprises:
the dividing processing unit is used for dividing the historical voltage data set into a training set and a test set according to a preset proportion, wherein the test set is the voltage data set to be detected;
and the distribution network voltage prediction model construction unit is used for training the Bi-LSTM neural network prediction model by using the training set to construct a distribution network voltage prediction model.
Further, the voltage error value obtaining module is configured to determine the plurality of voltage error values by the following formula:
st＝|yt-xt|
wherein s istIndicating the value of the voltage error at time t, ytIndicates the predicted value of voltage, x, at time ttIndicating the actual voltage measurement at time t.
Further, the voltage error standard deviation value obtaining module is configured to determine the voltage error standard deviation value according to the following formula:
wherein μ represents an average value of the voltage error values, m represents the number of the voltage error values, t represents a time, stRepresents the voltage error value at time t, and σ represents the standard deviation value of the voltage error.
The present application has been described in detail with reference to specific embodiments and illustrative examples, but the description is not intended to limit the application. Those skilled in the art will appreciate that various equivalent substitutions, modifications or improvements may be made to the presently disclosed embodiments and implementations thereof without departing from the spirit and scope of the present disclosure, and these fall within the scope of the present disclosure. The protection scope of this application is subject to the appended claims.
Claims (10)
1. A distribution network voltage abnormal data detection method is characterized by comprising the following steps:
acquiring historical distribution network voltage data and generating a historical voltage data set;
acquiring a voltage data set to be detected according to the historical voltage data set, wherein the voltage data set to be detected comprises a plurality of voltage actual measurement values which are sequentially arranged according to a measurement time sequence;
inputting the voltage data set to be detected into a pre-constructed distribution network voltage prediction model, and obtaining a plurality of voltage prediction values output by the distribution network voltage prediction model, wherein the voltage prediction values are respectively in one-to-one correspondence with voltage actual measurement values in the voltage data set to be detected except for the first two voltage actual measurement values, and the distribution network voltage prediction model comprises a preset rule;
obtaining a plurality of voltage error values, wherein any voltage error value is a difference value between any predicted voltage value and a corresponding actual voltage measurement value;
determining a voltage error standard deviation value according to the voltage error values;
and determining whether the actual measured value of the target voltage is abnormal data or not according to the target voltage error value and the standard difference value of the voltage error, wherein the target voltage error value corresponds to the actual measured value of the target voltage, and the target voltage error value is any one of the voltage error values.
2. The method for detecting the abnormal data of the voltage of the distribution network according to claim 1, wherein the preset rules comprise corresponding input rules and output rules:
the input rule is that the actual voltage measurement value at the t-2 moment and the actual voltage measurement value at the t-1 moment in the voltage data set to be detected are used as the input of the distribution network voltage prediction model;
the output rule is that the actual voltage measurement value at the t-th moment in the voltage data set to be detected is used as the output of the distribution network voltage prediction model, and the t-th moment is the moment corresponding to any actual voltage measurement value except the corresponding moments of the two previous actual voltage measurement values in the voltage data set to be detected.
3. The method for detecting abnormal data of voltage of distribution network according to claim 1, wherein the step of constructing the prediction model of voltage of distribution network comprises:
dividing the historical voltage data set into a training set and a test set according to a preset proportion, wherein the test set is the voltage data set to be detected;
and training the Bi-LSTM neural network prediction model by using the training set to construct a distribution network voltage prediction model.
4. The method of detecting abnormal data of voltage in a distribution network according to claim 1, wherein the obtaining a plurality of voltage error values comprises:
determining the plurality of voltage error values by:
st＝|yt-xt|；
wherein s istIndicating the value of the voltage error at time t, ytIndicates the predicted value of voltage, x, at time ttIndicating the actual voltage measurement at time t.
5. The method of detecting abnormal data of voltage of a distribution network according to claim 1, wherein the determining a standard deviation value of voltage errors according to the plurality of voltage error values comprises:
determining the standard deviation value of the voltage error by the following formula:
wherein μ represents an average of the plurality of voltage error values,m represents the number of voltage error values, t represents the time, stRepresents the voltage error value at time t, and σ represents the standard deviation value of the voltage error.
6. A distribution network voltage abnormal data detection device is characterized in that the distribution network voltage abnormal data detection device is applied to the distribution network voltage abnormal data detection method of any one of claims 1 to 5, and the distribution network voltage abnormal data detection device comprises:
the historical voltage data acquisition module is used for acquiring historical distribution network voltage data and generating a historical voltage data set;
the voltage data acquisition module is used for acquiring a voltage data set to be detected according to the historical voltage data set, wherein the voltage data set to be detected comprises a plurality of voltage actual measurement values which are sequentially arranged according to a measurement time sequence;
the voltage prediction value acquisition module is used for inputting the voltage data set to be detected into a pre-constructed distribution network voltage prediction model and acquiring a plurality of voltage prediction values output by the distribution network voltage prediction model, wherein the plurality of voltage prediction values are respectively in one-to-one correspondence with voltage actual measurement values in the voltage data set to be detected except for the first two voltage actual measurement values, and the distribution network voltage prediction model comprises a preset rule;
the voltage error value acquisition module is used for acquiring a plurality of voltage error values, wherein any voltage error value is a difference value between any predicted voltage value and a corresponding actual voltage measurement value;
the voltage error standard difference value acquisition module is used for determining a voltage error standard difference value according to the plurality of voltage error values;
and the abnormal data judgment module is used for determining whether the actual measured value of the target voltage is abnormal data according to a target voltage error value and the standard voltage error difference value, wherein the target voltage error value corresponds to the actual measured value of the target voltage, and the target voltage error value is any one of the voltage error values.
7. The distribution network voltage anomaly data detection device according to claim 6, wherein the preset rules include corresponding input rules and output rules:
the input rule is that the actual voltage measurement value at the t-2 moment and the actual voltage measurement value at the t-1 moment in the voltage data set to be detected are used as the input of the distribution network voltage prediction model;
the output rule is that the actual voltage measurement value at the t-th moment in the voltage data set to be detected is used as the output of the distribution network voltage prediction model, and the t-th moment is the moment corresponding to any actual voltage measurement value except the corresponding moments of the two previous actual voltage measurement values in the voltage data set to be detected.
8. The distribution network voltage abnormal data detection device according to claim 6, further comprising a model construction module for constructing the distribution network voltage prediction model in advance, wherein the model construction module comprises:
the dividing processing unit is used for dividing the historical voltage data set into a training set and a test set according to a preset proportion, wherein the test set is the voltage data set to be detected;
and the distribution network voltage prediction model construction unit is used for training the Bi-LSTM neural network prediction model by using the training set to construct a distribution network voltage prediction model.
9. The distribution network voltage anomaly data detection device according to claim 6, wherein the voltage error value acquisition module is configured to determine the plurality of voltage error values by the following formula:
st＝|yt-xt|；
wherein s istIndicating the value of the voltage error at time t, ytIndicates the predicted value of voltage, x, at time ttIndicating the actual voltage measurement at time t.
10. The distribution network voltage anomaly data detection device according to claim 6, wherein the voltage error standard deviation value obtaining module is configured to determine the voltage error standard deviation value according to the following formula:
wherein μ represents an average value of the voltage error values, m represents the number of the voltage error values, t represents a time, stRepresents the voltage error value at time t, and σ represents the standard deviation value of the voltage error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110214662.9A CN112819373A (en) | 2021-02-25 | 2021-02-25 | Distribution network voltage abnormal data detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110214662.9A CN112819373A (en) | 2021-02-25 | 2021-02-25 | Distribution network voltage abnormal data detection method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112819373A true CN112819373A (en) | 2021-05-18 |
Family
ID=75863931
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110214662.9A Pending CN112819373A (en) | 2021-02-25 | 2021-02-25 | Distribution network voltage abnormal data detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112819373A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113300371A (en) * | 2021-05-31 | 2021-08-24 | 广东电网有限责任公司 | Method and device for determining real-time voltage of power distribution room |
| CN113554229A (en) * | 2021-07-23 | 2021-10-26 | 国网青海省电力公司信息通信公司 | Three-phase voltage unbalance abnormality detection method and device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108089962A (en) * | 2017-11-13 | 2018-05-29 | 北京奇艺世纪科技有限公司 | A kind of method for detecting abnormality, device and electronic equipment |
| CN109033037A (en) * | 2018-07-26 | 2018-12-18 | 厦门大学 | Buoy automatic monitoring system data quality control method |
| CN109492866A (en) * | 2018-10-08 | 2019-03-19 | 南昌科晨电力试验研究有限公司 | A kind of distribution Running State intelligent evaluation method |
| CN110458230A (en) * | 2019-08-12 | 2019-11-15 | 江苏方天电力技术有限公司 | A kind of distribution transforming based on the fusion of more criterions is with adopting data exception discriminating method |
| US20200104639A1 (en) * | 2018-09-28 | 2020-04-02 | Applied Materials, Inc. | Long short-term memory anomaly detection for multi-sensor equipment monitoring |
| CN111340647A (en) * | 2019-11-21 | 2020-06-26 | 中国电力科学研究院有限公司 | Power distribution network data adjustment method and system |
| CN111608899A (en) * | 2020-04-28 | 2020-09-01 | 茂盟（上海）工程技术股份有限公司 | Water pump running state abnormity discrimination method based on efficiency analysis |
-
2021
- 2021-02-25 CN CN202110214662.9A patent/CN112819373A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108089962A (en) * | 2017-11-13 | 2018-05-29 | 北京奇艺世纪科技有限公司 | A kind of method for detecting abnormality, device and electronic equipment |
| CN109033037A (en) * | 2018-07-26 | 2018-12-18 | 厦门大学 | Buoy automatic monitoring system data quality control method |
| US20200104639A1 (en) * | 2018-09-28 | 2020-04-02 | Applied Materials, Inc. | Long short-term memory anomaly detection for multi-sensor equipment monitoring |
| CN109492866A (en) * | 2018-10-08 | 2019-03-19 | 南昌科晨电力试验研究有限公司 | A kind of distribution Running State intelligent evaluation method |
| CN110458230A (en) * | 2019-08-12 | 2019-11-15 | 江苏方天电力技术有限公司 | A kind of distribution transforming based on the fusion of more criterions is with adopting data exception discriminating method |
| CN111340647A (en) * | 2019-11-21 | 2020-06-26 | 中国电力科学研究院有限公司 | Power distribution network data adjustment method and system |
| CN111608899A (en) * | 2020-04-28 | 2020-09-01 | 茂盟（上海）工程技术股份有限公司 | Water pump running state abnormity discrimination method based on efficiency analysis |
Non-Patent Citations (1)
| Title |
|---|
| 周德才，孙亦鸣: "《计算机随机模拟原理、方法及计算程序》", 华中理工大学出版社, pages: 14 - 15 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113300371A (en) * | 2021-05-31 | 2021-08-24 | 广东电网有限责任公司 | Method and device for determining real-time voltage of power distribution room |
| CN113554229A (en) * | 2021-07-23 | 2021-10-26 | 国网青海省电力公司信息通信公司 | Three-phase voltage unbalance abnormality detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2808797A1 (en) | Operation management device, operation management method, and program | |
| EP2963553B1 (en) | System analysis device and system analysis method | |
| CN112819373A (en) | Distribution network voltage abnormal data detection method and device | |
| CN110929934A (en) | Equipment failure prediction method and device, computer equipment and storage medium | |
| CN112149860A (en) | Automatic anomaly detection method and system | |
| CN110715678B (en) | Sensor abnormity detection method and device | |
| CN111027531A (en) | Pointer instrument information identification method and device and electronic equipment | |
| CN105279386A (en) | Method and device for determining index abnormal data | |
| CN117394337A (en) | Power grid load early warning method and system thereof | |
| CN110769452A (en) | Method, system, server and storage medium for identifying longitude and latitude abnormity of base station | |
| CN104714429A (en) | Coal sample experiment data acquisition and analysis system and method | |
| Grbac et al. | Stability of software defect prediction in relation to levels of data imbalance | |
| CN114492629A (en) | Abnormality detection method, abnormality detection device, electronic apparatus, and storage medium | |
| CN113948159A (en) | Fault detection method, device and equipment for transformer | |
| CN107247871B (en) | Item detection time checking method for early warning and device | |
| CN113934536A (en) | Data acquisition method facing edge calculation | |
| CN116485079B (en) | Digital lean carbon emission accounting method, device, equipment and storage medium | |
| CN116520068B (en) | Diagnostic method, device, equipment and storage medium for electric power data | |
| CN115691144B (en) | Abnormal traffic state monitoring method, device and equipment and readable storage medium | |
| CN110956330A (en) | Method and system for predicting line loss of power transmission line based on multidimensional influence quantity | |
| CN114019946B (en) | Method and device for processing monitoring data of industrial control terminal | |
| CN115774159A (en) | Fault detection system for power unit of high-voltage frequency converter | |
| US20220187262A1 (en) | Device and method for anomaly detection of gas sensor | |
| CN115453447A (en) | Online detection method for out-of-tolerance electric meter based on suspected electric meter stepwise compensation rejection | |
| Łęczycki et al. | Extended sensor reliability evaluation method in multi-sensor control systems |
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 |