CN115128345B - Power grid safety early warning method and system based on harmonic monitoring - Google Patents
Power grid safety early warning method and system based on harmonic monitoring Download PDFInfo
- Publication number
- CN115128345B CN115128345B CN202210773252.2A CN202210773252A CN115128345B CN 115128345 B CN115128345 B CN 115128345B CN 202210773252 A CN202210773252 A CN 202210773252A CN 115128345 B CN115128345 B CN 115128345B
- Authority
- CN
- China
- Prior art keywords
- harmonic
- harmonic current
- frequency
- early warning
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R23/00—Arrangements for measuring frequencies; Arrangements for analysing frequency spectra
- G01R23/16—Spectrum analysis; Fourier analysis
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- General Physics & Mathematics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention provides a power grid safety early warning method and system based on harmonic monitoring, which are applied to the technical field of power grid safety early warning, and the method comprises the following steps: the harmonic signal time sequence information is obtained through the power quality monitoring module, and the harmonic current time sequence information is extracted from the harmonic signal time sequence information. And inputting the harmonic current time sequence information into an early warning characteristic analysis model to obtain the harmonic current early warning frequency characteristic. And carrying out region segmentation on the power grid region to be monitored, and obtaining a region segmentation result. And traversing the region segmentation result, and extracting region equipment harmonic characteristic information which comprises harmonic current deflection frequency. And positioning equipment according to the harmonic current early warning frequency characteristic and the harmonic current deflection frequency to determine harmonic abnormal equipment. And carrying out harmonic abnormality early warning according to the harmonic current early warning frequency characteristic and harmonic abnormality equipment. The method solves the technical problems of high false alarm probability and low early warning accuracy in the prior art in the power grid safety early warning based on harmonic monitoring.
Description
Technical Field
The invention relates to the technical field of power grid safety early warning, in particular to a power grid safety early warning method and system based on harmonic monitoring.
Background
The harmonic source of the power system is electrical equipment, the generation of the harmonic can cause harm to the whole electricity utilization environment, and in order to ensure the stable operation of the electricity utilization environment, the harmonic source needs to be pre-warned and positioned, and the influence of the harmonic source is reduced. However, the harmonic monitoring-based power grid safety early warning in the prior art has higher probability of false alarm due to the existence of circuit noise, and further has poor early warning accuracy.
Therefore, the power grid safety early warning based on harmonic monitoring in the prior art has the technical problems of high false warning probability and low early warning accuracy.
Disclosure of Invention
The utility model provides a power grid safety early warning method and system based on harmonic monitoring, which are used for solving the technical problems of high false alarm probability and low early warning accuracy of the power grid safety early warning based on harmonic monitoring in the prior art.
In view of the above problems, the present application provides a method and a system for power grid safety pre-warning based on harmonic monitoring.
In a first aspect of the present application, a method for power grid safety pre-warning based on harmonic monitoring is provided, where the method is applied to a power grid safety pre-warning system based on harmonic monitoring, and the system includes a distributed power quality monitoring module, and the method includes: uploading harmonic signal time sequence information of a public connection point of a power grid area to be monitored through a distributed power quality monitoring module; extracting harmonic current time sequence information from the harmonic signal time sequence information of any public connection point; inputting the harmonic current time sequence information into an early warning feature analysis model to obtain early warning feature information, wherein the early warning feature information comprises harmonic current early warning frequency features; performing region segmentation on the power grid region to be monitored according to the public connection point to obtain a region segmentation result; traversing the region segmentation result, and extracting regional equipment harmonic characteristic information, wherein the regional equipment harmonic characteristic information comprises harmonic current deflection frequency; positioning in the region segmentation result according to the harmonic current early warning frequency characteristic and the harmonic current deflection frequency to determine harmonic abnormal equipment; and carrying out harmonic abnormality early warning according to the harmonic current early warning frequency characteristic and the harmonic abnormality equipment.
In a second aspect of the present application, there is provided a power grid security pre-warning system based on harmonic monitoring, the system including a distributed power quality monitoring module, the system including: the harmonic signal time sequence information acquisition module is used for uploading harmonic signal time sequence information of a public connection point of a power grid area to be monitored through the distributed power quality monitoring module; the harmonic current time sequence information extraction module is used for extracting harmonic current time sequence information from the harmonic signal time sequence information of any public connection point; the early warning characteristic information acquisition module is used for inputting the harmonic current time sequence information into an early warning characteristic analysis model to acquire early warning characteristic information, wherein the early warning characteristic information comprises harmonic current early warning frequency characteristics; the region segmentation module is used for carrying out region segmentation on the power grid region to be monitored according to the public connection point to obtain a region segmentation result; the regional equipment harmonic characteristic information acquisition module is used for traversing the regional segmentation result and extracting regional equipment harmonic characteristic information, wherein the regional equipment harmonic characteristic information comprises harmonic current deflection frequency; the harmonic abnormal equipment determining module is used for positioning in the region segmentation result according to the harmonic current early-warning frequency characteristic and the harmonic current deflection frequency to determine harmonic abnormal equipment; and the abnormality early warning module is used for carrying out harmonic abnormality early warning according to the harmonic current early warning frequency characteristic and the harmonic abnormality equipment.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
according to the method provided by the embodiment of the application, the harmonic signal time sequence information is acquired, the harmonic current time sequence information is extracted, the harmonic current time sequence information is input into the early warning feature analysis model, early warning evaluation is performed on the harmonic current time sequence information, early warning grade evaluation is performed on harmonic abnormal frequency with abnormal content deviation degree or duration deviation degree, and the harmonic current early warning frequency feature is acquired, namely the abnormal harmonic abnormal frequency and the corresponding harmonic abnormal frequency early warning grade are acquired. And then acquiring harmonic characteristic information of equipment in the region according to the region segmentation result to acquire the deflection frequency of the harmonic current. Harmonic anomaly devices are determined according to harmonic current early warning frequency characteristics and harmonic current deflection frequencies. And carrying out harmonic abnormality pre-warning through harmonic current pre-warning frequency characteristics and harmonic abnormality equipment. Because the intelligent early warning characteristic analysis model is adopted, the harmonic signals are separated, the harmonic current characteristic information is acquired more accurately, the accurate evaluation of the time sequence information of the harmonic signals is realized, and the technical effect of early warning judgment accuracy is further improved. And further, the technical problems of high false alarm probability and low early warning accuracy in the prior art in the power grid safety early warning based on harmonic monitoring are solved.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Fig. 1 is a schematic flow chart of a power grid safety early warning method based on harmonic monitoring provided by the application;
fig. 2 is a schematic flow chart of acquiring early warning feature information in a power grid safety early warning method based on harmonic monitoring provided by the application;
fig. 3 is a schematic flow chart of a method for obtaining harmonic abnormal equipment in a power grid safety pre-warning method based on harmonic monitoring provided by the application;
fig. 4 is a schematic structural diagram of a power grid safety early warning system based on harmonic monitoring.
Reference numerals illustrate: the system comprises a harmonic signal time sequence information acquisition module 11, a harmonic current time sequence information extraction module 12, an early warning characteristic information acquisition module 13, a region segmentation module 14, a region equipment harmonic characteristic information acquisition module 15, a harmonic abnormal equipment determination module 16 and an abnormal early warning module 17.
Detailed Description
The utility model provides a power grid safety early warning method and system based on harmonic monitoring, which are used for solving the technical problems of high false alarm probability and low early warning accuracy of the power grid safety early warning based on harmonic monitoring in the prior art.
The technical solutions in the present application will be clearly and completely described below with reference to the accompanying drawings. The described embodiments are only some of the contents that can be realized by the present application, and not the whole contents of the present application.
Example 1
As shown in fig. 1, the present application provides a power grid safety pre-warning method based on harmonic monitoring, where the method is applied to a power grid safety pre-warning system based on harmonic monitoring, and the system includes a distributed power quality monitoring module, and the method includes:
step 100: uploading harmonic signal time sequence information of a public connection point of a power grid area to be monitored through a distributed power quality monitoring module;
step 200: extracting harmonic current time sequence information from the harmonic signal time sequence information of any public connection point;
specifically, the distributed power quality monitoring module is a monitoring device distributed at a public connection point of a power grid area to be detected, and is used for monitoring power quality parameters, such as display frequency, voltage fluctuation and flicker, voltage deviation, voltage fundamental wave effective value and true effective value, current fundamental wave effective value and true effective value, fundamental wave active power, fundamental wave apparent power, true power factor and the like, generated in a power distribution network in real time, wherein the power grid area to be detected comprises a power supply network area and a user network area. Wherein the common connection point is the connection point of the user network area and the power supply network area. For example, a plurality of users are connected on a certain power supply line, and the connection part of each user for entering the home is a public connection point. And acquiring harmonic signal time sequence information of a user network area and harmonic signal time sequence information of an area of a power supply network, which are obtained by monitoring waveform information of harmonic wave changes along with time, generated by each acquired parameter, of a public connection point through a distributed power quality monitoring module. And extracting harmonic current time sequence information from the harmonic signal time sequence information of any public connection point, wherein the harmonic current time sequence information is harmonic waveform information generated by the change of an actually monitored current signal along with time, namely the harmonic current waveform information.
Step 300: inputting the harmonic current time sequence information into an early warning feature analysis model to obtain early warning feature information, wherein the early warning feature information comprises harmonic current early warning frequency features;
specifically, the obtained harmonic current time sequence information, namely current harmonic waveform information, is input into an early warning feature analysis model, wherein the early warning feature analysis model is used for analyzing the harmonic current time sequence information, carrying out harmonic current frequency separation, harmonic current abnormality evaluation and harmonic current early warning grade evaluation on the harmonic current time sequence information, and obtaining early warning feature information. The early warning characteristic information comprises harmonic current early warning frequency characteristics, the harmonic current early warning frequency characteristics are obtained after the harmonic abnormal frequency characteristics are calibrated according to harmonic abnormal frequency early warning grades, and the harmonic current early warning frequency characteristics comprise the frequency of early warning harmonic current, namely the frequency of the harmonic current, and the corresponding early warning grades.
As shown in fig. 2, the method step 300 provided in the embodiment of the present application further includes:
step 310: acquiring a harmonic current frequency separation module, a harmonic current abnormality evaluation module and a harmonic current early-warning level evaluation module according to the early-warning characteristic analysis model;
Step 320: inputting the harmonic current time sequence information into the harmonic current frequency separation module to obtain harmonic current frequency distribution information, wherein the harmonic current frequency distribution information comprises a harmonic current frequency value, a frequency value content rate and a frequency value duration;
step 330: inputting the harmonic current frequency value, the frequency value content and the frequency value duration into the harmonic current abnormality evaluation module to obtain harmonic abnormal frequency, wherein the content deviation degree or the duration deviation degree;
step 340: inputting the content deviation degree, the duration deviation degree and the harmonic abnormal frequency into the harmonic current early-warning level evaluation module to obtain a harmonic abnormal frequency early-warning level;
step 350: and calibrating the harmonic abnormal frequency characteristic according to the harmonic abnormal frequency early warning level, generating the harmonic current early warning frequency characteristic, and adding the harmonic current early warning frequency characteristic into the early warning characteristic information.
Specifically, according to the early warning characteristic analysis model, a harmonic current frequency separation module, a harmonic current abnormality evaluation module and a harmonic current early warning grade evaluation module are obtained. The harmonic current frequency separation module performs frequency separation on the acquired harmonic current time sequence information to acquire harmonic current frequency distribution information, wherein the harmonic current frequency distribution information comprises a harmonic current frequency value, a frequency value content rate and a frequency value duration. And then, inputting a harmonic current frequency value, a frequency value content rate and a frequency value duration into the harmonic current abnormality evaluation module to obtain a harmonic abnormal frequency, a content rate deviation degree or a duration deviation degree, and inputting the content rate deviation degree and the duration deviation degree and the harmonic abnormal frequency into the harmonic current early warning grade evaluation module to obtain a harmonic abnormal frequency early warning grade. And finally, calibrating the harmonic abnormal frequency characteristic according to the harmonic abnormal frequency early-warning level, calibrating the harmonic abnormal frequency early-warning level, generating the harmonic current early-warning frequency characteristic, and adding the harmonic current early-warning frequency characteristic into the early-warning characteristic information. The harmonic abnormal frequency and the corresponding early warning level existing in the harmonic are determined by acquiring the harmonic current early warning frequency characteristics, and technical support is provided for the determination of subsequent abnormal equipment.
The method step 310 provided in the embodiment of the present application further includes:
step 311: acquiring harmonic current monitoring record data, wherein the harmonic current monitoring record data is real-time harmonic current monitoring data of a history record;
step 312: acquiring a harmonic current frequency record value, calibrating a characteristic value of the harmonic current monitoring record data according to the harmonic current frequency record value, and acquiring a harmonic current frequency separation module identification data set;
step 313: and training the harmonic current frequency separation module according to the harmonic current monitoring record data and the harmonic current frequency separation module identification data set.
Specifically, the harmonic current frequency separation module obtains harmonic current monitoring record data, wherein the harmonic current monitoring record data is real-time harmonic current monitoring data of a history record. And processing the harmonic current monitoring record data, namely the real-time harmonic current monitoring data of the history record, in a Fourier transform or wavelet transform mode to obtain frequency data of the harmonic current in the harmonic current frequency record value, and the frequency value content and the frequency value duration record value data together form a harmonic current frequency record value. And then calibrating the characteristic value of the harmonic current monitoring record data according to the harmonic current frequency record value, calibrating the frequency data of the historical record real-time harmonic current monitoring data in the harmonic current monitoring record data, the frequency value content and the frequency value duration, and obtaining a harmonic current frequency separation module identification data set. And then constructing a decision tree by taking the harmonic current monitoring record data and the harmonic current frequency separation module identification data set as input data to finish the construction of the harmonic current frequency separation module. The harmonic current frequency separation module is used for performing frequency separation on harmonic current to obtain frequency data, frequency value content and frequency value duration of each component harmonic.
The method provided in the embodiment of the present application further includes step 313:
step 313-1: constructing a first decision tree according to the harmonic current monitoring record data and the harmonic current frequency separation module identification data set;
step 313-2: extracting a training data set which does not meet the preset accuracy rate in the first decision tree, and marking the training data set as a first data set;
step 313-3: and when the data volume of the first data set is smaller than or equal to a preset data volume, setting the first decision tree as the harmonic current frequency separation module.
Specifically, a first decision tree is constructed according to the harmonic current monitoring record data as training data and the harmonic current frequency separation module identification data set as supervision data. The decision tree is a prediction model, and the acquisition of the harmonic current frequency record value frequency can be completed by inputting harmonic current time sequence information. The decision tree is a supervised learning process in which the harmonic current frequency separation module identifies a dataset as the supervised data for the decision tree. And acquiring a training data set which does not meet the preset accuracy in the first decision tree, and recording the training data set as a first data set. The training data set which does not meet the preset accuracy rate is data which is obtained by constructing a decision tree through the training data set and has deviation between an obtained result and the supervision data exceeding a certain threshold, and at the moment, the data in the first data set cannot obtain an accurate result in the first decision tree. When the data volume of the first data set is smaller than or equal to the preset data volume, the output result of the first decision tree obtained at the moment and the requirement of accurately processing most data can be met, and the first decision tree is set as a harmonic current frequency separation module at the moment. The obtained harmonic current frequency separation module outputs a result more accurately by obtaining the first data set and judging the data quantity in the first data set.
The method step 313-3 provided in the embodiment of the present application further includes:
step 313-31: when the data volume of the first data set is larger than the preset data volume, constructing a second decision tree through the first data set;
step 313-32: extracting a training data set which does not meet the preset accuracy rate in the second decision tree, and marking the training data set as a second data set;
steps 313-33: and repeating iteration based on the second data set until the data volume of the N data set of the N decision tree is smaller than or equal to the preset data volume, merging the first decision tree to the N decision tree, and generating the harmonic current frequency separation module.
Specifically, when the data volume of the first data set is larger than the preset data volume, there is more data that cannot acquire an accurate output result in the first decision tree, and at this time, the second decision tree is continuously constructed by taking the first data set as input data. And when the second decision tree is constructed, the supervision data of the second decision tree is the data in the identification data set of the harmonic current frequency separation module corresponding to the first data set. And continuously extracting a training data set which does not meet the preset accuracy in a second decision tree, recording the training data set as a second data set, continuously repeating iteration based on the second data set until the data amount of the N data set of the N decision tree is smaller than or equal to the preset data amount, merging the first decision tree to the N decision tree, and generating the harmonic current frequency separation module. And combining the N decision tree until the data volume of the N data set is smaller than or equal to the preset data volume, so that the obtained harmonic current frequency separation module can finish accurate processing of the data, and the model can accurately obtain the abnormal frequency of the harmonic, the deviation degree of the content or the deviation degree of the duration.
The method provided in the embodiment of the present application further includes step 340:
step 341: acquiring a harmonic current frequency content threshold and a frequency duration threshold according to the harmonic current abnormality evaluation module;
step 342: judging whether the frequency value content meets the harmonic current frequency content threshold value or not;
step 343: if not, judging whether the duration of the frequency value meets the threshold of the duration of the frequency value;
step 344: if yes, adding the harmonic current frequency value into the harmonic abnormal frequency, and calculating the duration deviation according to the duration of the frequency value and the duration threshold.
Specifically, according to the harmonic current anomaly evaluation module, a harmonic current frequency content threshold and a frequency duration threshold are obtained, wherein the harmonic current frequency content threshold is a threshold of a ratio of a root mean square value of a harmonic current to a root mean square value of a fundamental current, namely a threshold of a harmonic current content rate, the frequency duration threshold is a threshold of a duration of a current frequency, and the harmonic current frequency content threshold and the frequency duration threshold can be set according to practical situations. And then judging whether the frequency value content in the harmonic current frequency distribution information meets the harmonic current frequency content threshold value or not, namely judging whether the frequency value content is larger than or equal to the harmonic current frequency content threshold value or not, and adding the harmonic current frequency value into the harmonic abnormal frequency if the influence of the harmonic on the original fundamental wave is larger and the influence on the current waveform of the power grid is larger when the frequency value content is larger than or equal to the harmonic current frequency content threshold value. When the frequency content is not satisfied, that is, when the frequency content is smaller than the wave current frequency content threshold, the influence on the original fundamental wave is small. And then judging whether the duration of the frequency value meets the duration threshold of the frequency, namely judging whether the duration of the current frequency is larger than or equal to the duration threshold of the frequency. When the duration threshold value of the current frequency is larger than or equal to the duration threshold value of the frequency, the influence duration of the harmonic on the original fundamental wave is longer, and if the current frequency is abnormal, otherwise, no abnormality exists. And adding harmonic current frequency meeting a frequency duration threshold into the harmonic abnormal frequency, and calculating the duration deviation according to the frequency value duration and the frequency duration threshold, namely acquiring a deviation value between the frequency value duration and the frequency duration threshold so as to judge the influence degree of the harmonic current frequency value.
The method step 342 provided in the embodiment of the present application further includes:
step 342-1: when the frequency value content rate meets the harmonic current frequency content threshold value, adding the harmonic current frequency value into the harmonic abnormal frequency, and calculating the content rate deviation according to the frequency value content rate and the harmonic current frequency content threshold value.
And if the frequency value content meets the harmonic current frequency content threshold, adding the harmonic current frequency value into the harmonic abnormal frequency, and calculating the content deviation according to the frequency value content and the harmonic current frequency content threshold, namely acquiring a deviation value between the frequency value content and the harmonic current frequency content threshold so as to judge the influence degree of the harmonic current frequency value.
The method step 310 provided in the embodiment of the present application further includes:
step 314: constructing a first coordinate axis based on big data, wherein the first coordinate axis is used for carrying out early warning rating on the content deviation;
step 315: constructing a second coordinate axis based on big data, wherein the second coordinate axis is used for carrying out early warning rating on the duration deviation degree;
Step 316: generating an early warning grade evaluation coordinate system according to the first coordinate axis and the second coordinate axis;
step 317: and setting the early warning level evaluation coordinate system as the harmonic current early warning level evaluation module.
Specifically, a first coordinate axis is constructed based on big data, the first coordinate axis is the deviation of the content rate, and the mapping relation between the deviation of the content rate and the early warning rating is obtained through the big data, namely, each deviation of the content rate corresponds to one early warning rating. And constructing a second coordinate axis based on the big data, wherein the relation between the duration deviation degree and the early warning rating is obtained through the big data, namely, each duration deviation degree corresponds to one early warning rating, and the second coordinate axis is the duration deviation degree. And generating an early warning grade evaluation coordinate system according to the relation between the first coordinate axis and the second coordinate axis and the corresponding early warning grade. And in the early warning level evaluation coordinate system, the abscissa is a first coordinate axis, the ordinate is a second coordinate axis, each coordinate point corresponds to two early warning levels of the early warning level of the content deviation degree and the early warning level of the duration deviation degree, and the final calculation result is obtained by carrying out mean value calculation on the two early warning levels, namely the actual early warning level of the point. And setting the early warning level evaluation coordinate system as the harmonic current early warning level evaluation module. The harmonic current early warning level evaluation module is obtained through big data, so that the harmonic current early warning level evaluation module has universality and the obtained early warning level evaluation result is more accurate.
Step 400: performing region segmentation on the power grid region to be monitored according to the public connection point to obtain a region segmentation result;
step 500: traversing the region segmentation result, and extracting regional equipment harmonic characteristic information, wherein the regional equipment harmonic characteristic information comprises harmonic current deflection frequency;
specifically, region segmentation is performed on the power grid region to be detected according to the public connection point, a monitoring region of the power quality monitoring module at each node is obtained, and a region segmentation result is obtained. And traversing all the region segmentation results, extracting harmonic characteristic information of the devices in the region, and when the harmonic characteristic information of the devices in the region is obtained, obtaining the characteristic information of harmonic current frequency values generated in the working state of the characterization device confirmed according to the historical monitoring data, namely obtaining the waveform information of the harmonic current generated in the historical working state of the working device generating the harmonic in the region. The harmonic characteristic information of the regional equipment comprises harmonic current deflection frequency, wherein the harmonic current deflection frequency is a harmonic current frequency value of which the occurrence frequency is larger than a preset value in a historical working state of the equipment, namely, the harmonic frequency with the longest harmonic duration generated in the historical working state is obtained.
Step 600: positioning in the region segmentation result according to the harmonic current early warning frequency characteristic and the harmonic current deflection frequency to determine harmonic abnormal equipment;
step 700: and carrying out harmonic abnormality early warning according to the harmonic current early warning frequency characteristic and the harmonic abnormality equipment.
Specifically, according to the obtained harmonic current early warning frequency characteristics, namely harmonic current frequency and harmonic current deflection frequency of early warning in the harmonic signal time sequence information obtained by each public connection point, namely harmonic frequency with longest harmonic duration generated by the equipment in a working state, equipment positioning is carried out in an area segmentation result of an area to be detected, and harmonic abnormal equipment is determined. The harmonic current early warning frequency characteristics comprise the frequency of early warning harmonic current, namely the frequency of harmonic current, and the corresponding early warning level. And determining the early warning grade of the equipment according to the determined harmonic abnormal equipment and the harmonic current early warning frequency characteristics, and carrying out harmonic abnormal early warning on the equipment according to the early warning grade.
As shown in fig. 3, the method step 600 provided in the embodiment of the present application further includes:
step 610: matching according to the deviation frequency of the harmonic current and the characteristic of the early warning frequency of the harmonic current to obtain initial harmonic abnormal equipment;
Step 620: detecting harmonic current values of the initial harmonic abnormal equipment along the harmonic current direction information through the distributed power quality monitoring module to obtain a plurality of groups of input harmonic current values and output harmonic current values, wherein the initial harmonic abnormal equipment corresponds to the plurality of groups of input harmonic current values and output harmonic current values one by one;
step 630: comparing any one of the input harmonic current values with the output harmonic current value;
step 640: adding the initial harmonic abnormal equipment with the output harmonic current value larger than the input harmonic current value into a list of equipment to be screened;
step 650: and adding the harmonic abnormal equipment according to the equipment in the equipment list to be screened.
Specifically, the harmonic current deviation frequency is matched with the harmonic current early warning frequency characteristic, namely the harmonic current deviation frequency obtained according to the equipment historical operation data is matched with the harmonic current early warning frequency characteristic, and initial harmonic abnormal equipment is obtained. And then carrying out harmonic current value detection on the initial harmonic abnormal equipment along the harmonic current direction information through the distributed power quality monitoring module to obtain a plurality of groups of input harmonic current values and output harmonic current values, wherein the initial harmonic abnormal equipment corresponds to the plurality of groups of input harmonic current values and output harmonic current values one by one. And comparing any group of input harmonic current values with the output harmonic current values, and when the output harmonic current values are larger than the input harmonic current values, directly generating harmonic waves by the equipment at the moment, so that harmonic anomalies exist in the equipment. At this time, the initial harmonic anomaly device whose output harmonic current value is greater than the input harmonic current value is added to a list of devices to be screened. And adding the harmonic abnormal equipment according to the equipment in the equipment list to be screened. By monitoring the harmonic current value of the operation equipment, the harmonic abnormal equipment in the region segmentation result is accurately determined.
In summary, the method provided by the embodiment of the application acquires the harmonic signal time sequence information, extracts the harmonic current time sequence information, inputs the harmonic current time sequence information into the early warning feature analysis model, performs early warning evaluation on the harmonic current time sequence information, performs early warning grade evaluation on the harmonic abnormal frequency with abnormal deviation degree of the content or duration, and acquires the harmonic current early warning frequency feature, namely acquires the abnormal harmonic abnormal frequency with abnormal and the corresponding harmonic abnormal frequency early warning grade. And then acquiring harmonic characteristic information of equipment in the region according to the region segmentation result to acquire the deflection frequency of the harmonic current. Harmonic anomaly devices are determined according to harmonic current early warning frequency characteristics and harmonic current deflection frequencies. And carrying out harmonic anomaly early warning through harmonic anomaly frequency early warning levels and harmonic anomaly equipment. Because the intelligent early warning characteristic analysis model is adopted, the harmonic signals are separated, the harmonic current characteristic information is acquired more accurately, the accurate evaluation of the time sequence information of the harmonic signals is realized, and the technical effect of early warning judgment accuracy is further improved. And further, the technical problems of high false alarm probability and low early warning accuracy in the prior art in the power grid safety early warning based on harmonic monitoring are solved.
Example two
Based on the same inventive concept as the power grid safety pre-warning method based on harmonic monitoring in the foregoing embodiments, as shown in fig. 4, the present application provides a power grid safety pre-warning system based on harmonic monitoring, where the system includes a distributed power quality monitoring module, and the system includes:
the harmonic signal time sequence information acquisition module 11 is used for uploading harmonic signal time sequence information of a public connection point of a power grid area to be monitored through the distributed power quality monitoring module;
a harmonic current timing information extraction module 12, configured to extract harmonic current timing information from the harmonic signal timing information of any one of the common connection points;
the early warning characteristic information acquisition module 13 is used for inputting the harmonic current time sequence information into an early warning characteristic analysis model to obtain early warning characteristic information, wherein the early warning characteristic information comprises harmonic current early warning frequency characteristics;
the region segmentation module 14 is configured to segment the power grid region to be monitored according to the common connection point, and obtain a region segmentation result;
the regional equipment harmonic characteristic information acquisition module 15 is used for traversing the regional segmentation result and extracting regional equipment harmonic characteristic information, wherein the regional equipment harmonic characteristic information comprises harmonic current deflection frequency;
A harmonic abnormal equipment determining module 16, configured to determine a harmonic abnormal equipment by positioning in the region segmentation result according to the harmonic current early-warning frequency characteristic and the harmonic current deviation frequency;
and the abnormality early warning module 17 is used for carrying out harmonic abnormality early warning according to the harmonic current early warning frequency characteristic and the harmonic abnormality equipment.
Further, the early warning feature information acquisition module 13 is further configured to:
acquiring a harmonic current frequency separation module, a harmonic current abnormality evaluation module and a harmonic current early-warning level evaluation module according to the early-warning characteristic analysis model;
inputting the harmonic current time sequence information into the harmonic current frequency separation module to obtain harmonic current frequency distribution information, wherein the harmonic current frequency distribution information comprises a harmonic current frequency value, a frequency value content rate and a frequency value duration;
inputting the harmonic current frequency value, the frequency value content and the frequency value duration into the harmonic current abnormality evaluation module to obtain harmonic abnormal frequency, wherein the content deviation degree or the duration deviation degree;
inputting the content deviation degree, the duration deviation degree and the harmonic abnormal frequency into the harmonic current early-warning level evaluation module to obtain a harmonic abnormal frequency early-warning level;
And calibrating the harmonic abnormal frequency characteristic according to the harmonic abnormal frequency early warning level, generating the harmonic current early warning frequency characteristic, and adding the harmonic current early warning frequency characteristic into the early warning characteristic information.
Further, the early warning feature information acquisition module 13 is further configured to:
acquiring harmonic current monitoring record data, wherein the harmonic current monitoring record data is real-time harmonic current monitoring data of a history record;
acquiring a harmonic current frequency record value, calibrating a characteristic value of the harmonic current monitoring record data according to the harmonic current frequency record value, and acquiring a harmonic current frequency separation module identification data set;
and training the harmonic current frequency separation module according to the harmonic current monitoring record data and the harmonic current frequency separation module identification data set.
Further, the early warning feature information acquisition module 13 is further configured to:
constructing a first decision tree according to the harmonic current monitoring record data and the harmonic current frequency separation module identification data set;
extracting a training data set which does not meet the preset accuracy rate in the first decision tree, and marking the training data set as a first data set;
and when the data volume of the first data set is smaller than or equal to a preset data volume, setting the first decision tree as the harmonic current frequency separation module.
Further, the early warning feature information acquisition module 13 is further configured to:
when the data volume of the first data set is larger than the preset data volume, constructing a second decision tree through the first data set;
extracting a training data set which does not meet the preset accuracy rate in the second decision tree, and marking the training data set as a second data set;
and repeating iteration based on the second data set until the data volume of the N data set of the N decision tree is smaller than or equal to the preset data volume, merging the first decision tree to the N decision tree, and generating the harmonic current frequency separation module.
Further, the early warning feature information acquisition module 13 is further configured to:
acquiring a harmonic current frequency content threshold and a frequency duration threshold according to the harmonic current abnormality evaluation module;
judging whether the frequency value content meets the harmonic current frequency content threshold value or not;
if not, judging whether the duration of the frequency value meets the threshold of the duration of the frequency value;
if yes, adding the harmonic current frequency value into the harmonic abnormal frequency, and calculating the duration deviation according to the duration of the frequency value and the duration threshold.
Further, the early warning feature information acquisition module 13 is further configured to:
when the frequency value content rate meets the harmonic current frequency content threshold value, adding the harmonic current frequency value into the harmonic abnormal frequency, and calculating the content rate deviation according to the frequency value content rate and the harmonic current frequency content threshold value.
Further, the early warning feature information acquisition module 13 is further configured to:
constructing a first coordinate axis based on big data, wherein the first coordinate axis is used for carrying out early warning rating on the content deviation;
constructing a second coordinate axis based on big data, wherein the second coordinate axis is used for carrying out early warning rating on the duration deviation degree;
generating an early warning grade evaluation coordinate system according to the first coordinate axis and the second coordinate axis;
and setting the early warning level evaluation coordinate system as the harmonic current early warning level evaluation module.
Further, the harmonic anomaly device determination module 16 is further configured to:
matching according to the deviation frequency of the harmonic current and the characteristic of the early warning frequency of the harmonic current to obtain initial harmonic abnormal equipment;
detecting harmonic current values of the initial harmonic abnormal equipment along the harmonic current direction information through the distributed power quality monitoring module to obtain a plurality of groups of input harmonic current values and output harmonic current values, wherein the initial harmonic abnormal equipment corresponds to the plurality of groups of input harmonic current values and output harmonic current values one by one;
Comparing any one of the input harmonic current values with the output harmonic current value;
adding the initial harmonic abnormal equipment with the output harmonic current value larger than the input harmonic current value into a list of equipment to be screened;
and adding the harmonic abnormal equipment according to the equipment in the equipment list to be screened.
The second embodiment is used for executing the method as in the first embodiment, and the execution principle and the execution basis thereof can be obtained through the content described in the first embodiment, which is not repeated herein. Although the present application has been described in connection with specific features and embodiments thereof, the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application, and the content thus obtained also falls within the scope of the present application.
Claims (10)
1. The utility model provides a power grid safety precaution method based on harmonic monitoring, its characterized in that, the method is applied to a power grid safety precaution system based on harmonic monitoring, and the system includes distributed electric energy quality monitoring module, and the method includes:
uploading harmonic signal time sequence information of a public connection point of a power grid area to be monitored through a distributed power quality monitoring module;
Extracting harmonic current time sequence information from the harmonic signal time sequence information of any public connection point;
inputting the harmonic current time sequence information into an early warning feature analysis model to obtain early warning feature information, wherein the early warning feature information comprises harmonic current early warning frequency features;
performing region segmentation on the power grid region to be monitored according to the public connection point to obtain a region segmentation result;
traversing the region segmentation result, and extracting regional equipment harmonic characteristic information, wherein the regional equipment harmonic characteristic information comprises harmonic current deflection frequency;
positioning in the region segmentation result according to the harmonic current early warning frequency characteristic and the harmonic current deflection frequency to determine harmonic abnormal equipment;
and carrying out harmonic abnormality early warning according to the harmonic current early warning frequency characteristic and the harmonic abnormality equipment.
2. The method of claim 1, wherein inputting the harmonic current timing information into an early warning feature analysis model to obtain early warning feature information comprises:
acquiring a harmonic current frequency separation module, a harmonic current abnormality evaluation module and a harmonic current early-warning level evaluation module according to the early-warning characteristic analysis model;
Inputting the harmonic current time sequence information into the harmonic current frequency separation module to obtain harmonic current frequency distribution information, wherein the harmonic current frequency distribution information comprises a harmonic current frequency value, a frequency value content rate and a frequency value duration;
inputting the harmonic current frequency value, the frequency value content and the frequency value duration into the harmonic current abnormality evaluation module to obtain harmonic abnormal frequency, content deviation and duration deviation;
inputting the content deviation degree, the duration deviation degree and the harmonic abnormal frequency into the harmonic current early-warning level evaluation module to obtain a harmonic abnormal frequency early-warning level;
and calibrating the harmonic abnormal frequency characteristic according to the harmonic abnormal frequency early warning level, generating the harmonic current early warning frequency characteristic, and adding the harmonic current early warning frequency characteristic into the early warning characteristic information.
3. The method of claim 2, wherein the acquiring harmonic current frequency separation module comprises:
acquiring harmonic current monitoring record data, wherein the harmonic current monitoring record data is real-time harmonic current monitoring data of a history record;
Acquiring a harmonic current frequency record value, calibrating a characteristic value of the harmonic current monitoring record data according to the harmonic current frequency record value, and acquiring a harmonic current frequency separation module identification data set;
and training the harmonic current frequency separation module according to the harmonic current monitoring record data and the harmonic current frequency separation module identification data set.
4. The method of claim 3, wherein said training the harmonic current frequency separation module based on the harmonic current monitoring record data and the harmonic current frequency separation module identification data set comprises:
constructing a first decision tree according to the harmonic current monitoring record data and the harmonic current frequency separation module identification data set;
extracting a training data set which does not meet the preset accuracy rate in the first decision tree, and marking the training data set as a first data set;
and when the data volume of the first data set is smaller than or equal to a preset data volume, setting the first decision tree as the harmonic current frequency separation module.
5. The method of claim 4, wherein the method further comprises:
when the data volume of the first data set is larger than the preset data volume, constructing a second decision tree through the first data set;
Extracting a training data set which does not meet the preset accuracy rate in the second decision tree, and marking the training data set as a second data set;
and repeating iteration based on the second data set until the data volume of the N data set of the N decision tree is smaller than or equal to the preset data volume, merging the first decision tree to the N decision tree, and generating the harmonic current frequency separation module.
6. The method of claim 2, wherein inputting the harmonic current anomaly evaluation module with the harmonic current frequency value, the frequency value content and the frequency value duration to obtain a harmonic anomaly frequency and content deviation and duration deviation comprises:
acquiring a harmonic current frequency content threshold and a frequency duration threshold according to the harmonic current abnormality evaluation module;
judging whether the frequency value content meets the harmonic current frequency content threshold value or not;
if the harmonic current frequency content threshold is not met, continuing to judge whether the frequency value duration meets the frequency duration threshold or not;
and if the frequency duration threshold is met, adding the harmonic current frequency value into the harmonic abnormal frequency, and calculating the duration deviation degree according to the frequency value duration and the frequency duration threshold.
7. The method of claim 6, wherein said determining whether said frequency value content meets said harmonic current frequency content threshold comprises:
when the frequency value content rate meets the harmonic current frequency content threshold value, adding the harmonic current frequency value into the harmonic abnormal frequency, and calculating the content rate deviation according to the frequency value content rate and the harmonic current frequency content threshold value.
8. The method of claim 2, wherein obtaining the harmonic current warning level assessment module comprises:
constructing a first coordinate axis based on big data, wherein the first coordinate axis is used for carrying out early warning rating on the content deviation;
constructing a second coordinate axis based on big data, wherein the second coordinate axis is used for carrying out early warning rating on the duration deviation degree;
generating an early warning grade evaluation coordinate system according to the first coordinate axis and the second coordinate axis;
and setting the early warning level evaluation coordinate system as the harmonic current early warning level evaluation module.
9. The method of claim 1, wherein the locating in the region segmentation result based on the harmonic current warning frequency characteristic and the harmonic current bias frequency, determining a harmonic anomaly device, comprises:
Matching according to the deviation frequency of the harmonic current and the characteristic of the early warning frequency of the harmonic current to obtain initial harmonic abnormal equipment;
detecting harmonic current values of the initial harmonic abnormal equipment along the harmonic current direction information through the distributed power quality monitoring module to obtain a plurality of groups of input harmonic current values and output harmonic current values, wherein the initial harmonic abnormal equipment corresponds to the plurality of groups of input harmonic current values and output harmonic current values one by one;
comparing any one of the input harmonic current values with the output harmonic current value;
adding the initial harmonic abnormal equipment with the output harmonic current value larger than the input harmonic current value into a list of equipment to be screened;
and adding the equipment in the equipment list to be screened into the harmonic abnormal equipment.
10. A power grid security early warning system based on harmonic monitoring, the system comprising a distributed power quality monitoring module, the system comprising:
the harmonic signal time sequence information acquisition module is used for uploading harmonic signal time sequence information of a public connection point of a power grid area to be monitored through the distributed power quality monitoring module;
The harmonic current time sequence information extraction module is used for extracting harmonic current time sequence information from the harmonic signal time sequence information of any public connection point;
the early warning characteristic information acquisition module is used for inputting the harmonic current time sequence information into an early warning characteristic analysis model to acquire early warning characteristic information, wherein the early warning characteristic information comprises harmonic current early warning frequency characteristics;
the region segmentation module is used for carrying out region segmentation on the power grid region to be monitored according to the public connection point to obtain a region segmentation result;
the regional equipment harmonic characteristic information acquisition module is used for traversing the regional segmentation result and extracting regional equipment harmonic characteristic information, wherein the regional equipment harmonic characteristic information comprises harmonic current deflection frequency;
the harmonic abnormal equipment determining module is used for positioning in the region segmentation result according to the harmonic current early-warning frequency characteristic and the harmonic current deflection frequency to determine harmonic abnormal equipment;
and the abnormality early warning module is used for carrying out harmonic abnormality early warning according to the harmonic current early warning frequency characteristic and the harmonic abnormality equipment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210773252.2A CN115128345B (en) | 2022-07-01 | 2022-07-01 | Power grid safety early warning method and system based on harmonic monitoring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210773252.2A CN115128345B (en) | 2022-07-01 | 2022-07-01 | Power grid safety early warning method and system based on harmonic monitoring |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115128345A CN115128345A (en) | 2022-09-30 |
| CN115128345B true CN115128345B (en) | 2023-04-28 |
Family
ID=83381872
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210773252.2A Active CN115128345B (en) | 2022-07-01 | 2022-07-01 | Power grid safety early warning method and system based on harmonic monitoring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115128345B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115795315B (en) * | 2022-11-03 | 2023-05-12 | 北京市永康药业有限公司 | Abnormality sensing method and system for chemical liquid supply equipment |
| CN116207845B (en) * | 2022-11-28 | 2024-05-28 | 无锡广盈集团有限公司 | Automatic monitoring method and system for protecting power equipment |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4937205B2 (en) * | 2008-07-22 | 2012-05-23 | 株式会社東芝 | Power quality evaluation system |
| CN106841859B (en) * | 2017-01-06 | 2019-08-09 | 云南电网有限责任公司电力科学研究院 | Electric energy quality monitoring method for early warning and system based on characteristic Optimum Matching |
| CN109613324B (en) * | 2018-11-07 | 2021-10-08 | 全球能源互联网研究院有限公司 | Harmonic amplification detection method and device |
| CN110472195B (en) * | 2019-08-15 | 2023-04-18 | 国网福建省电力有限公司莆田供电公司 | User harmonic hazard early warning and assessment method based on interval monitoring data |
| CN110633870A (en) * | 2019-09-24 | 2019-12-31 | 国家电网有限公司 | Harmonic early warning method, harmonic early warning device and terminal equipment |
-
2022
- 2022-07-01 CN CN202210773252.2A patent/CN115128345B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN115128345A (en) | 2022-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115128345B (en) | Power grid safety early warning method and system based on harmonic monitoring | |
| Zheng et al. | Raw wind data preprocessing: A data-mining approach | |
| CN107480705B (en) | Tap water pipe water leakage detection method | |
| CN109375060B (en) | Method for calculating fault waveform similarity of power distribution network | |
| CN113124929A (en) | Transformer substation multi-parameter signal acquisition comprehensive analysis system and method | |
| Chen et al. | Fault anomaly detection of synchronous machine winding based on isolation forest and impulse frequency response analysis | |
| CN105974223B (en) | A kind of method and system for on-line checking electrical equipment working condition | |
| CN112504511A (en) | Generator stator temperature monitoring method, device and medium | |
| CN114978877B (en) | Abnormality processing method, abnormality processing device, electronic equipment and computer readable medium | |
| CN114325256A (en) | Power equipment partial discharge identification method, system, equipment and storage medium | |
| CN108508297A (en) | A kind of fault arc detection method based on change coefficient and SVM | |
| Damala et al. | A simple decision tree-based disturbance monitoring system for VSC-based HVDC transmission link integrating a DFIG wind farm | |
| CN117572837B (en) | Intelligent power plant AI active operation and maintenance method and system | |
| CN117368651B (en) | Comprehensive analysis system and method for faults of power distribution network | |
| Ghorbaniparvar et al. | Bootstrap-based hypothesis test for detecting sustained oscillations | |
| CN113884837A (en) | Online monitoring and analyzing system and method for partial discharge of cable | |
| CN109635430A (en) | Grid power transmission route transient signal monitoring method and system | |
| CN114184870A (en) | Non-invasive load identification method and equipment | |
| CN115343579B (en) | Power grid fault analysis method and device and electronic equipment | |
| CN108761250B (en) | Industrial control equipment voltage and current-based intrusion detection method | |
| CN112731208B (en) | Low-voltage line fault and abnormity on-line monitoring method, equipment and medium | |
| CN115146715A (en) | Power utilization potential safety hazard diagnosis method, device, equipment and storage medium | |
| Chen et al. | Identification of typical partial discharge defects of distribution system equipment based on classification learner | |
| Zhao et al. | Research on a vibration signal analysis method for motor bearing | |
| US11200134B2 (en) | Anomaly detection apparatus, method, and program recording medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |