CN117949884B - Power quality and voltage monitoring and calibrating method and system based on machine learning - Google Patents
Power quality and voltage monitoring and calibrating method and system based on machine learning Download PDFInfo
- Publication number
- CN117949884B CN117949884B CN202410348644.3A CN202410348644A CN117949884B CN 117949884 B CN117949884 B CN 117949884B CN 202410348644 A CN202410348644 A CN 202410348644A CN 117949884 B CN117949884 B CN 117949884B
- Authority
- CN
- China
- Prior art keywords
- module
- calibration
- data
- power grid
- machine learning
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 75
- 238000010801 machine learning Methods 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 claims abstract description 91
- 238000005070 sampling Methods 0.000 claims abstract description 59
- 238000012545 processing Methods 0.000 claims abstract description 57
- 238000011156 evaluation Methods 0.000 claims abstract description 34
- 230000005856 abnormality Effects 0.000 claims abstract description 18
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 16
- 230000000694 effects Effects 0.000 claims description 15
- 230000002159 abnormal effect Effects 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000012952 Resampling Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 abstract description 5
- 238000013441 quality evaluation Methods 0.000 abstract description 2
- 238000005259 measurement Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013486 operation strategy Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/005—Calibrating; Standards or reference devices, e.g. voltage or resistance standards, "golden" references
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the technical field of measuring electric variables, and provides a machine learning-based electric energy quality and voltage monitoring and calibrating method and system, wherein the method comprises the following steps: s1, acquiring the current moment as a sampling time t, performing electric energy quality and voltage monitoring through a detection module, and capturing relevant data of a power grid at the sampling time t; s2, processing related data of the power grid through a data processing module; s3, extracting key features from the related data of the power grid processed by the data processing module by the feature extracting module; s4, the machine learning module processes the key features to form an evaluation result; s5, calculating a state index ACT of power quality and voltage monitoring according to the evaluation result by the abnormality monitoring module; s6, automatically calibrating the detection module. The machine learning module outputs the scores of the power grid lines by applying the machine learning algorithm, so that the reliability and the accuracy of the power quality evaluation of the power grid lines are improved.
Description
Technical Field
The invention relates to the technical field of measuring electric variables, in particular to a machine learning-based electric energy quality and voltage monitoring and calibrating method and system.
Background
Under the condition that the operation of the current power grid is increasingly complex, the electric energy quality monitoring system is widely applied, a large amount of analysis data is provided for power grid detection, and an immeasurable effect is achieved for the stable operation of the power grid.
As disclosed in chinese patent CN116184100B, a method and an apparatus for calibrating power quality of a power grid, along with development of a power system, the power quality of each node of the whole power grid is different; the power quality consists of a plurality of quality contents, and each power quality needs to be comprehensively evaluated to know the power quality condition of the whole power grid; along with the marketization of electric power and the transparency of electric energy quality, a plurality of different electric energy quality indexes and the electric energy quality of each node should be comprehensively analyzed so as to meet the needs of electric power users. But cannot be actively and automatically estimated, which is unfavorable for large-scale electric energy quality and voltage monitoring.
In addition, current technology relies primarily on conventional power quality monitoring instruments and equipment that can record key parameters of the power grid in real time, but their functionality may be limited when dealing with complex power issues such as short-time voltage fluctuations, flicker, harmonics, etc. Meanwhile, the conventional calibration method is usually performed manually, requiring a professional technician and complicated steps, which are time-consuming and may cause human errors.
Meanwhile, in the existing power quality and voltage monitoring process of the power grid, limitation may exist when a large amount of or complex data is processed, the manual calibration process is time-consuming and error-prone, and the response speed and accuracy of the whole system are affected.
The invention is designed for solving the problems that the assessment is inaccurate, the active assessment is impossible, the abnormal state cannot be early-warned, the interaction capability is poor, the intelligence degree is low and the like in the prior art.
Disclosure of Invention
The invention aims to provide a machine learning-based power quality and voltage monitoring and calibrating method for overcoming the defects existing at present.
In order to overcome the defects in the prior art, the invention adopts the following technical scheme:
a power quality and voltage monitoring and calibrating method based on machine learning comprises the following steps:
s1, acquiring the current moment as a sampling time t, performing electric energy quality and voltage monitoring through a detection module deployed at a key node of a power grid, capturing relevant data of the power grid at the sampling time t, and storing the relevant data of the power grid;
s2, processing the related data of the power grid through a data processing module, and eliminating noise and invalid data so as to ensure the quality of the related data of the power grid;
S3, extracting key features from the related data of the power grid processed by the data processing module by the feature extracting module;
S4, the machine learning module processes the key features to form an evaluation result;
S5, the abnormality monitoring module calculates a state index ACT of power quality and voltage monitoring according to the evaluation result, and sends a calibration request signal to the calibration module after the state index ACT is obtained;
and S6, after receiving the calibration request signal, the calibration module automatically calibrates the detection module.
Optionally, the step of processing the related data by the data processing module in step S2 includes:
s21, collecting relevant data of an original power grid from a detection module;
s22, carrying out noise processing on the related data of the original power grid;
s23, converting the related data of the power grid subjected to noise processing into a standard format so that the data is in a universal range;
S24, resampling the related data of the power grid converted into the standard mode, ensuring uniformity of the time sequence, and dividing continuous data into windows with fixed sizes for local analysis and feature extraction.
Optionally, extracting the key features in step S3 includes the steps of:
S31, waveform data in voltage signals or current signals of key nodes are obtained from the related data processed by the data processing module, and frequency component characteristics, harmonic distortion characteristics, phase angle characteristics, power factor characteristics and load fluctuation characteristics of the voltage signals or the current signals are obtained;
S32, constructing a feature vector according to the frequency component features, the harmonic distortion features, the phase angle features, the power factor features and the load fluctuation features;
Optionally, in step S3, the frequency component characteristic, the harmonic distortion characteristic, the phase angle characteristic, the power factor characteristic, and the load fluctuation characteristic of the key node at the sampling time t are obtained for normalization, and a characteristic vector FeatureVector is formed:
;
wherein ω ' is normalized phase angle characteristic, PE ' is normalized power factor characteristic, load_f ' is normalized Load fluctuation characteristic, F ' is normalized frequency component characteristic, THD ' is normalized total harmonic distortion characteristic.
Optionally, in step S4, the processing by the machine learning module of the key feature to form an evaluation result includes the following steps:
S41, outputting a Score of the key node at the sampling time t by using the machine learning module, wherein the machine learning module is trained by using the history vector feature to obtain a trained machine learning module, and the history feature vector is obtained by processing the power grid history related data sequentially through the data processing module and the feature extraction module.
Optionally, the anomaly monitoring module calculates the state index ACT of the power grid transmission line at the sampling time t according to the Score of the key node at the sampling time t output by the machine learning module:
;
Wherein w i is the weight of the ith key node of the power grid, alpha is an adjustment weight, the value of the adjustment weight is set according to practical conditions, T is the monitored time length before T sampling time, S i (T) is the score of the ith key node of the power grid at the sampling time T, D (T 0) is a time attenuation factor, S i(t0) is the score of the ith key node of the power grid at the T 0 sampling time in T, and N is the number of key nodes;
after the state index ACT is calculated, the calibration request signal is generated, and the calibration request signal is sent to the calibration module.
Optionally, in step S6, the calibration module performs automatic calibration on the detection module according to the following steps after receiving the calibration request signal:
S61, after receiving a calibration request signal sent by the abnormality monitoring module, the calibration module compares the calculated state index ACT with a preset or historical reference value to determine whether calibration is needed;
s62, if the absolute value of the difference between the state index and the reference value is larger than a preset value, the calibration module sends a calibration signal to the detection module and jumps to S64, otherwise, the calibration is not needed, and the step S63 is jumped to;
S63, the detection module does not need to be adjusted, and the step S1 is returned;
S64, the detection module adjusts filter setting and sampling rate based on the calibration signal, and jumps to step S65;
s65, after adjustment is completed, the current moment is obtained again to serve as a sampling time t, the steps S1-S62 are repeatedly executed, the calibration effect is verified through the step S62, whether the current detection module still needs to be calibrated is judged, if the calibration effect meets the condition that the absolute value of the difference between the state index and the reference value is larger than a preset value, the step S64 is skipped, and if the calibration effect does not meet the condition that the absolute value of the difference between the state index and the reference value is larger than the preset value, the step S63 is skipped.
Optionally, the normalized Load fluctuation feature load_f' is calculated according to the following formula:
;
Wherein min (load_f) is the minimum value of the Load fluctuation feature in the history data of the ith key node, max (load_f) is the maximum value of the Load fluctuation feature in the history data of the ith key node, load_f is the Load fluctuation feature, and the value is calculated according to the following formula:
;
Max (Load) and min (Load) are the maximum value and the minimum value of the Load in the observation period respectively, and Average (Load) is the Average value of the Load in the historical data of the key nodes of the power grid.
The power quality and voltage monitoring and calibrating system based on machine learning is applied to a power quality and voltage monitoring and calibrating method based on machine learning, and comprises a server, a detection module, a data processing module, a feature extraction module, a machine learning module, an anomaly monitoring module and a calibrating module, wherein the server stores intermediate data and process data of the detection module, the data processing module, the feature extraction module, the machine learning module, the anomaly monitoring module and the calibrating module;
The detection module acquires relevant data of key nodes on a power grid line, the data processing module is used for processing the relevant data so as to ensure the quality of the relevant data, the feature extraction module extracts key features from the processed relevant data, the machine learning module evaluates the key features to form an evaluation result, the abnormality monitoring module calculates a state index ACT of power quality and voltage monitoring according to the evaluation result and judges whether the state index ACT is abnormal, and the calibration module automatically calibrates the detection module;
the key features include frequency component features, harmonic distortion features, phase angle features, power factor features, and load fluctuation features, among others.
Optionally, the detection module includes a voltage detection unit, a current detection unit and a data storage, the voltage detection unit collects voltage data of key nodes of the power grid, the current detection unit collects current data of the key nodes of the power grid, and the data storage stores the voltage data and the current data.
Optionally, the calibration module includes a calibration unit and a prompt unit, the prompt unit is used for prompting the abnormality of the power quality and the voltage monitoring to the manager, and the calibration unit is used for calibrating the detection module.
The beneficial effects obtained by the invention are as follows:
1. The extracted key features are evaluated through an anomaly evaluation module, so that the accurate evaluation of the detection state of the power grid line is improved, and the reliability and the accuracy of electric energy quality and voltage monitoring are ensured;
2. The machine learning module is used for outputting the scores of the power grid lines by applying a machine learning algorithm, so that the reliability and the accuracy of the power quality evaluation of the power grid lines are improved, the active evaluation of abnormal states is also considered, and the whole system is guaranteed to have the advantages of rich evaluation means, reliable evaluation quality and strong interaction capability;
3. Through the cooperation between calibration module and the unusual monitoring module for abnormal state can be by accurate detection, and calibrate according to the unusual result, guaranteed that entire system has that early warning state evaluation is accurate, interactive ability is strong and evaluation ability is strong's advantage.
4. The data processing module and the detection module are matched with each other, so that the data obtained by power grid line detection can be processed, the processing precision of the whole system to the monitoring process is improved, and the whole system is guaranteed to have the advantages of high processing capacity and high intelligent degree.
Drawings
The invention will be further understood from the following description taken in conjunction with the accompanying drawings. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the embodiments. Like reference numerals designate like parts in the different views.
FIG. 1 is a schematic diagram of the calibration process according to the present invention.
Fig. 2 is a schematic flow chart of the invention for extracting key features.
Fig. 3 is a schematic diagram of an evaluation flow of the state index ACT according to the present invention.
FIG. 4 is a flow chart of the calibration operation of the calibration module according to the present invention.
Fig. 5 is a block schematic diagram of the power quality and voltage monitoring calibration system of the present invention.
Detailed Description
The following embodiments of the present invention are described in terms of specific examples, and those skilled in the art will appreciate the advantages and effects of the present invention from the disclosure herein. The invention is capable of other and different embodiments and its several details are capable of modification and variation in various respects, all without departing from the spirit of the present invention. The drawings of the present invention are merely schematic illustrations, and are not intended to be drawn to actual dimensions. The following embodiments will further illustrate the related art content of the present invention in detail, but the disclosure is not intended to limit the scope of the present invention.
Embodiment one: the embodiment provides a power quality and voltage monitoring and calibrating method based on machine learning, fig. 1 is a schematic diagram of a calibrating flow of the present invention, and as shown in fig. 1, the power quality and voltage monitoring and calibrating method includes the following steps:
S1, acquiring the current moment as a sampling time t, performing electric energy quality and voltage monitoring through a detection module deployed at a key node of a power grid, capturing relevant data of the power grid at the sampling time t, and storing the relevant data of the power grid; wherein the key nodes include, but are not limited to, the following list of several: the position of the expected monitoring, the input end of the transformer station, the output end of the transformer station, the load end and the like;
In addition, in the present embodiment, the related data includes, but is not limited to, several of the following: voltage magnitude, current intensity, frequency, phase angle, etc.;
s2, processing the related data of the power grid through a data processing module, and eliminating noise and invalid data so as to ensure the quality of the related data of the power grid;
S3, extracting key features from the related data of the power grid processed by the data processing module by the feature extracting module;
S4, the machine learning module processes the key features to form an evaluation result;
S5, the abnormality monitoring module calculates a state index ACT of power quality and voltage monitoring according to the evaluation result, and sends a calibration request signal to the calibration module after the state index ACT is obtained;
and S6, after receiving the calibration request signal, the calibration module automatically calibrates the detection module.
Optionally, the step of processing the related data by the data processing module in step S2 includes:
s21, collecting relevant data of an original power grid from a detection module;
s22, carrying out noise processing on the related data of the original power grid;
s23, converting the related data of the power grid subjected to noise processing into a standard format so that the data is in a universal range;
S24, resampling the related data of the power grid converted into the standard mode, ensuring uniformity of the time sequence, and dividing continuous data into windows with fixed sizes for local analysis and feature extraction.
Alternatively, fig. 2 is a schematic flow chart of the method for extracting key features. As shown in fig. 2, extracting key features at step S3 includes the steps of:
S31, waveform data in voltage signals or current signals of key nodes are obtained from the related data processed by the data processing module, and frequency component characteristics, harmonic distortion characteristics, phase angle characteristics, power factor characteristics and load fluctuation characteristics of the voltage signals or the current signals are obtained;
S32, constructing a feature vector according to the frequency component features, the harmonic distortion features, the phase angle features, the power factor features and the load fluctuation features;
Optionally, in step S3, the frequency component characteristic F (k), the harmonic distortion characteristic (Total Harmonic Distortion, abbreviated as THD), the phase angle characteristic, the power factor characteristic, and the load fluctuation characteristic of the key node at the sampling time t are obtained for normalization, and a characteristic vector FeatureVector is formed.
;
Wherein ω ' is normalized phase angle characteristic, PE ' is normalized power factor characteristic, load_f ' is normalized Load fluctuation characteristic, F ' is normalized frequency component characteristic, THD ' is normalized total harmonic distortion characteristic.
Wherein ω' is calculated according to the following formula:
;
Wherein max (ω) is the maximum value of the phase angle in the history data of the ith key node, min (ω) is the minimum value of the phase angle in the history data of the ith key node, ω is the phase angle of the key node at the sampling time t, and the value is directly obtained according to the phase difference between the voltage waveform and the current waveform, which is a technical means well known to those skilled in the art, so that a detailed description is omitted in this embodiment.
The normalized power factor characteristic PE' is calculated according to the following equation:
;
Wherein min (PE) is the minimum value of the power factor in the historical data of the ith key node, max (PE) is the maximum value of the power factor in the historical data of the ith key node, PE is the power factor of the ith key node at the sampling time t, and the calculation is carried out according to the following formula:
;
wherein ω is the phase angle of the critical node at the sampling time t, where the phase angle is the phase difference of the current waveform relative to the voltage waveform.
The normalized Load fluctuation characteristic load_f' is calculated according to the following equation:
;
Wherein min (load_f) is the minimum value of the Load fluctuation feature in the history data of the ith key node, max (load_f) is the maximum value of the Load fluctuation feature in the history data of the ith key node, load_f is the Load fluctuation feature, and the value is calculated according to the following formula:
;
Max (Load) and min (Load) are respectively the maximum value and the minimum value of the Load in the observation period, and Average (Load) is the Average value of the Load in the historical data of the key nodes of the power grid.
The normalized frequency component characteristic F' is calculated according to the following equation:
;
Wherein min (F (k)) is the minimum value of the frequency component in the history data of the i-th key node, max (F (k)) is the maximum value of the frequency component in the history data of the i-th key node, and F (k) is the k-th frequency component characteristic of the i-th key node, the values of which are calculated (using fourier transform FFT) according to the following formula:
;
Wherein x (n) is time series data of voltage or current signals, namely voltage or current signals acquired for the nth time; n is the total number of sampling points, i represents the ith critical node, and k represents the kth frequency component characteristic of the ith critical node.
In this embodiment, the monitoring of the frequency component characteristic F' may be considered by one of the current signal or the voltage signal, and the operator may select an appropriate index (such as the voltage signal or the current signal) according to the actual requirement.
The normalized total harmonic distortion characteristic THD' is calculated according to the following equation:
;
Wherein min (THD) is the minimum value of the total harmonic distortion in the historical data of the ith key node, max (THD) is the maximum value of the total harmonic distortion in the historical data of the ith key node, THD is the total harmonic distortion of the ith key node at the sampling time t, and the value is calculated according to the following formula:
;
Wherein V h is the voltage amplitude of the H harmonic, V 1 is the fundamental voltage amplitude, and H is the total harmonic frequency.
Optionally, in step S4, the processing by the machine learning module of the key feature to form an evaluation result includes the following steps:
S41, outputting a Score of the key node at the sampling time t by using the machine learning module, wherein the machine learning module is trained by using the history vector feature to obtain a trained machine learning module, and the history feature vector is obtained by processing the power grid history related data sequentially through the data processing module and the feature extraction module.
Fig. 3 is a schematic diagram of an evaluation flow of the state index ACT according to the present invention. As shown in fig. 3, determining the current time as a sampling time, performing data sampling based on the sampling time, inputting a feature vector FeatureVector into a machine learning module after performing data processing, and calculating a Score of a key node at a sampling time t; inputting the obtained Score of the key node at the sampling time t into an anomaly monitoring module, and calculating a state index ACT of a power grid transmission line at the sampling time t according to the Score of the key node at the sampling time t output by the machine learning module by the anomaly monitoring module; the calibration module judges whether calibration is needed, if not, the power quality and voltage monitoring task is executed based on the current detection module; if so, continuously judging whether the calibration times of the calibration module are more than 3 times, if not, triggering the calibration module to calibrate the monitoring of the detection module, updating the sampling time, and re-executing the evaluation flow of the whole state index ACT, if so, executing the power quality and voltage monitoring task based on the current detection module, and searching for problems in other aspects, thereby avoiding infinite repeated circulation.
In this embodiment, the operation can be realized by taking the feature vectors obtained by combination as the input of the machine learning module and outputting the Score of the key node at the sampling time t through the machine learning module through training.
Meanwhile, in the present embodiment, during the training of the machine learning module, the machine learning module performs automatic learning by using a machine learning algorithm and outputs the Score of the key node at the sampling time t.
Among other things, the machine learning algorithms described above include, but are not limited to, the following list of: such as decision trees, random forests, support vector machines, etc.
Meanwhile, historical data (including normal and abnormal conditions) is used for training, the historical data is input into a data processing module and a feature extraction module for processing, then the processed data is used for training a model of a machine learning module, and a cross-validation mode is adopted for validation, so that the accuracy and the reliability of the machine learning module are ensured.
When the machine learning module is qualified in training, the machine learning module is put into use, the feature vector FeatureVector extracted by the feature extraction module is input into the machine learning module, and the Score of the key node at the sampling time t is output.
In addition, the operation of training the machine learning module through the machine learning algorithm and outputting the Score of the key node at the sampling time t is a technical means well known to those skilled in the art, and those skilled in the art can query the related technical manual to learn the technology, so that the description is not repeated in this embodiment.
The scores of the key nodes are used for measuring the achievement degree of performance indexes or quality standards of the power supply quality.
Optionally, the anomaly monitoring module calculates the state index ACT of the power grid transmission line according to the Score of the key node at the sampling time t, which is output by the machine learning module:
;
Wherein w i is the weight of the ith key node of the power grid, α is an adjustment weight, the value is set according to practical situations, T is the monitored time length before T is the sampling time, S i (T) is the score of the ith key node of the power grid at the sampling time T, D (T 0) is a time attenuation factor, S i(t0) is the score of the ith key node of the power grid at the T 0 sampling time in T, N is the number of key nodes, and D (T) is a time attenuation factor, the value of which is calculated according to the following formula:
;
where λ is the decay rate, t is the sampling time, and its value is empirically obtained.
Wherein, in this example, the attenuation rate λ may be determined by:
1) Based on empirical selection:
if there is a priori knowledge or experience about the change in the grid conditions, a suitable decay rate can be estimated from this information.
For example, if it is known that the grid conditions typically change significantly within a few hours, a corresponding decay rate may be set, ensuring that data within this time frame has a greater impact on the current assessment.
2) The dynamic attenuation rate is adopted:
In some cases, the decay rate may be dynamically adjusted to accommodate changing environments based on current network conditions or other external factors.
For example: the system sets the decay rate to λ=0.1, and for the score at sample time 5, the time decay factor will be calculated as:
;
Wherein the decay factor D (t) may be applied to calculate the impact of past scores on the current score.
In this embodiment, the value range of the weight w i for the ith key node of the power grid is usually a positive number, and the value range may be 0 to 1, and in this embodiment, an example of value is provided:
1) If there are only three key nodes in a line:
For the main transformer substation, which is critical to the power grid, the weight of the position is as follows: w 1 = 0.5;
The large-scale industrial user has a larger load but is not as good as the key node of the main transformer substation, and the weight value is as follows: w 2 = 0.3;
for different residential areas, the load is relatively small, and the weight value is as follows: w 3 = 0.2.
2) In the urban grid scenario (5 key nodes):
The key nodes comprise: power station, main transformer station, industrial area, commercial area and residential area, then the value of weight is:
The power station is a main power source of the power grid, and the key effect of supplying power to the whole power grid can be given higher weight, and then the weight w 1 =0.4;
The main transformer substation is responsible for converting the power of the power station to a voltage suitable for urban distribution, which is also very critical, but slightly lower than the power station, and the weight is: w 2 = 0.3;
The large-scale industrial users of industry district, the electric power demand is big, and for power station and transformer substation, the importance is slightly low, then the weight is: w 3 = 0.15;
The business area includes a plurality of business buildings, such as shopping malls and office buildings, and the power demand is medium, the weights are: w 4 = 0.1;
the residential area contains numerous home users, the power demand is dispersed, and is relatively less important in the whole power grid, then the weight is: w 5 = 0.05.
In short, the weight distribution is set by the manager according to the actual condition of the key nodes, and meanwhile, the specific value of the weight is not fixed and is adjusted according to the actual condition and the operation strategy of the power grid.
In addition, the value range of the adjustment coefficient alpha is between 0 and 1, and is determined according to the following conditions:
When α=1, the state index is based entirely on the current score, irrespective of the historical data;
When α=0, the state index is based entirely on the historical score, ignoring the current score.
In summary, the value of choice α depends on the relative importance of the current score and the historical score in assessing the grid status.
If the current situation is considered more important, α should be set higher;
if the historical trend is more important for understanding the overall situation, α should be set lower.
In the present embodiment, an example is provided: the grid lines in a place have some fluctuations in the past few months (e.g. due to weather or equipment failure), but since the last month the grid has run very smooth.
In this case, a higher alpha value, such as 0.8, can be chosen to reflect the importance of the current steady state, with this alpha value (alpha=0.8), the impact of the current score is 80% and the impact of the historical score is only 20%.
Such weight assignment ensures that the state index ACT reflects more recent stability than past fluctuations.
In summary, the value of α may be different in different application scenarios, and this parameter should be determined according to the specific operation situation and the historical data of the power grid line.
Optionally, fig. 4 is a schematic flow chart of the calibration operation of the calibration module of the present invention, as shown in fig. 4, in step S6, the calibration module automatically calibrates the detection module according to the following steps:
S61, after receiving a calibration request signal sent by the abnormality monitoring module, the calibration module compares the calculated state index ACT with a preset or historical reference value to determine whether calibration is needed;
s62, if the absolute value of the difference between the state index and the reference value is larger than a preset value, the calibration module sends a calibration signal to the detection module and jumps to S64, otherwise, the calibration is not needed, and the step S63 is jumped to;
S63, the detection module does not need to be adjusted, and the step S1 is returned;
S64, the detection module adjusts filter setting and sampling rate based on the calibration signal, and jumps to step S65;
s65, after adjustment is completed, the current moment is obtained again to serve as a sampling time t, the steps S1-S62 are repeatedly executed, the calibration effect is verified through the step S62, whether the current detection module still needs to be calibrated is judged, if the calibration effect meets the condition that the absolute value of the difference between the state index and the reference value is larger than a preset value, the step S64 is skipped, and if the calibration effect does not meet the condition that the absolute value of the difference between the state index and the reference value is larger than the preset value, the step S63 is skipped.
In addition, the invention also provides a power quality and voltage monitoring and calibrating system based on machine learning, and fig. 5 is a block schematic diagram of the power quality and voltage monitoring and calibrating system according to the invention, wherein the power quality and voltage monitoring and calibrating system comprises a server, a detection module, a data processing module, a feature extraction module, a machine learning module, an abnormality monitoring module and a calibrating module, the server is respectively connected with the detection module, the data processing module, the feature extraction module, the machine learning module, the abnormality monitoring module and the calibrating module, and intermediate data and process data of the detection module, the data processing module, the feature extraction module, the machine learning module, the abnormality monitoring module and the calibrating module are stored in a database of the server for inquiry or calling.
The detection module acquires relevant data of key nodes on a power grid line, the data processing module is used for processing the relevant data so as to ensure the quality of the relevant data, the feature extraction module extracts key features from the processed relevant data, the machine learning module evaluates the key features to form an evaluation result, the abnormality monitoring module calculates a state index ACT of power quality and voltage monitoring according to the evaluation result and judges whether the state index ACT is abnormal, and the calibration module automatically calibrates the detection module.
Key features include, but are not limited to, the following list of several: frequency component characteristic F (k), harmonic distortion characteristic THD, phase angle characteristic, power factor characteristic, and load fluctuation characteristic.
Optionally, the detection module comprises a voltage detection unit, a current detection unit and a data memory, wherein the voltage detection unit collects voltage data of key nodes of the power grid, the current detection unit collects current data of the key nodes of the power grid, and the data memory stores the voltage data collected by the voltage detection unit and the current data collected by the current detection unit.
In this embodiment, after the voltage waveform and the current waveform of the power grid line are detected by the voltage detection unit and the current detection unit, the phase angle is calculated by a conventional manner, which is a technical means well known to those skilled in the art, so in this embodiment, a detailed description is omitted.
Optionally, the calibration module includes a calibration unit and a prompting unit, the prompting unit prompts the manager of the power quality and voltage monitoring abnormality, and the calibration unit is used for calibrating the detection module.
The prompting unit comprises a prompting device and a data transmitter, wherein the data transmitter detects abnormal data of the power quality and the voltage in the data transmission value prompting device, and the prompting device prompts abnormal data to a manager.
The calibration unit performs operations as in steps S61 to S65 during calibration of the detection module.
Specifically, the calibration unit automatically calibrates the detection module according to the following steps:
S61, after receiving a calibration request signal sent by the abnormality monitoring module, the calibration module compares the calculated state index ACT with a preset or historical reference value to determine whether calibration is needed;
s62, if the absolute value of the difference between the state index and the reference value is larger than a preset value, the calibration module sends a calibration signal to the detection module and jumps to S64, otherwise, the calibration is not needed, and the step S63 is jumped to;
S63, the detection module does not need to be adjusted, and the step S1 is returned;
S64, the detection module adjusts filter setting and sampling rate based on the calibration signal, and jumps to step S65;
s65, after adjustment is completed, the current moment is obtained again to serve as a sampling time t, the steps S1-S62 are repeatedly executed, the calibration effect is verified through the step S62, whether the current detection module still needs to be calibrated is judged, if the calibration effect meets the condition that the absolute value of the difference between the state index and the reference value is larger than a preset value, the step S64 is skipped, and if the calibration effect does not meet the condition that the absolute value of the difference between the state index and the reference value is larger than the preset value, the step S63 is skipped.
When the detection module receives the calibration signal, the filter setting is adjusted, and a higher-order filter is used to improve the filtering precision. And adjusting the sampling rate setting, and improving the sampling rate to improve the accuracy of sampling.
A lower ACT value indicates better performance or power quality, while a higher ACT value indicates possible problems or performance degradation.
Through the cooperation between calibration module and the unusual monitoring module for abnormal state can be by accurate detection, and calibrate according to the unusual result, guaranteed that entire system has that early warning state evaluation is accurate, interactive ability is strong and evaluation ability is strong's advantage.
Embodiment two: this embodiment should be understood to include all the features of any one of the foregoing embodiments, and further improve on the foregoing embodiments, where the power quality and voltage monitoring calibration system further includes a calibration evaluation module that evaluates the detection data of the calibrated detection module and dynamically adjusts the calibration operation of the calibration module on the detection module.
The calibration evaluation module obtains the reading of the detection module in a sampling period T after calibration, and calculates the qualification index QS of the detection module according to the following formula:
;
wherein, gamma 1、γ2、γ3 is a weight coefficient, the value of which is set according to the actual situation, AWS is an accuracy score, SAS is a stability score, PNFS is a performance compliance score, and the value of which is calculated according to the following formula:
;
Where R measured is the reading taken when the detection module performs the measurement after calibration of the measurement device, R expected is the ideal value of the measurement object of the detection module, which is a standard value predefined when setting, testing or calibrating the electrical device, i.e. this index is typically based on the design specifications of the device, industry standards, regulatory requirements or requirements of a specific application, such as a theoretical measurement value within the allowed deviation of the actual value from the industry standards.
The performance compliance score PNFS is calculated according to:
;
wherein σ is the standard deviation of the readings generated after calibration, the value of which is calculated according to the following equation:
;
Wherein i represents the ith measurement of the detection module in the calibration process, R i is the reading obtained by the ith measurement of the detection module in the calibration process, H is the total measurement times in the calibration process, mu is the arithmetic average value of all the readings after single calibration, and the value is calculated according to the following formula:
;
Where i represents the ith measurement of the detection module during the calibration process, R i is the reading obtained by the ith measurement of the detection module during the calibration process, and H is the total number of measurements performed during the calibration process.
The accuracy score AWS is calculated according to the following equation:
;
Wherein CV (measured) is the coefficient of variation of the measured value, PNFS is the performance compliance score, and satisfies:
;
where σ is the standard deviation of the readings generated after calibration, μ refers to the arithmetic mean of all the readings after a single calibration.
It should be noted that the calibration is performed at fixed intervals, such as: at intervals of 10 minutes or 1 hour.
The power quality and voltage monitoring calibration method comprises the following steps:
s71, a calibration evaluation module acquires the reading of the detection module in a sampling period after calibration, and calculates the qualification index QS of the detection module in the sampling period;
S72, the calibration evaluation module compares the calculated QS with a deviation range [ E 1,E2 ] allowed by the system or the manager, if the qualified index QS is met and does not fall into the deviation range [ E 1,E2 ] allowed by the system or the manager, the step S73 is skipped, and if the qualified index QS is met and falls into the deviation range [ E 1,E2 ] allowed by the system or the manager, the step S74 is skipped;
S73, the calibration module continues to calibrate the detection module, and the operations of the steps S71-S72 are repeatedly executed;
s74, the calibration state of the calibration module meets the requirement, and the calibration module stops calibrating the detection module.
The allowable deviation range [ E 1,E2 ] set by the system or the manager is set by the system or the manager according to the actual situation, which is a technical means known to those skilled in the art, and those skilled in the art can query the related technical manual to obtain the technology, so that the description is omitted in this embodiment.
Through the cooperation between calibration evaluation module and the detection module for the calibration process is more accurate, reliable, promotes the high efficiency and the reliability of whole calibration process.
The foregoing disclosure is only a preferred embodiment of the present invention and is not intended to limit the scope of the invention, so that all equivalent technical changes made by applying the description of the present invention and the accompanying drawings are included in the scope of the present invention, and in addition, elements in the present invention can be updated as the technology develops.
Claims (8)
1. The power quality and voltage monitoring and calibrating method based on machine learning is characterized by comprising the following steps of:
S1, acquiring the current moment as a sampling time t, performing electric energy quality and voltage monitoring through a detection module deployed at a key node of a power grid, capturing relevant data of the power grid at the sampling time t, and storing the relevant data of the power grid;
S2, processing the related data of the power grid through a data processing module, and eliminating noise and invalid data to ensure the quality of the related data of the power grid;
s3, extracting key features from the related data of the power grid processed by the data processing module by the feature extracting module, wherein the step S3 of extracting the key features comprises the following steps:
S31, waveform data in voltage signals or current signals of key nodes are obtained from the related data processed by the data processing module, and frequency component characteristics, harmonic distortion characteristics, phase angle characteristics, power factor characteristics and load fluctuation characteristics of the voltage signals or the current signals are obtained;
S32, constructing a feature vector according to the frequency component features, the harmonic distortion features, the phase angle features, the power factor features and the load fluctuation features;
S4, processing the key features by the machine learning module to form an evaluation result, wherein in the step S4, processing the key features by the machine learning module to form the evaluation result comprises the following steps:
S41, processing the feature vector by using a trained machine learning module to obtain a Score of the key node at a sampling time t, wherein the machine learning module is trained by using a historical feature vector to obtain the trained machine learning module, the historical feature vector is obtained by processing power grid historical related data sequentially through a data processing module and a feature extraction module, and a state index ACT of a power grid transmission line at the sampling time t is calculated:
;
Wherein w i is the weight of the ith key node of the power grid, alpha is an adjustment weight, the value of the adjustment weight is set by a manager according to actual conditions, T is the monitored time length before T sampling time, S i (T) is the score of the ith key node of the power grid at the sampling time T, D (T 0) is a time attenuation factor, S i(t0) is the score of the ith key node of the power grid at the T 0 sampling time in T, and N is the number of key nodes;
after the state index ACT is calculated, generating the calibration request signal, and sending the calibration request signal to a calibration module;
S5, an abnormality monitoring module calculates a state index ACT of power quality and voltage monitoring according to the evaluation result, and sends a calibration request signal to the calibration module after the state index ACT is obtained;
s6, the calibration module automatically calibrates the detection module after receiving the calibration request signal.
2. The machine learning based power quality and voltage monitoring calibration method according to claim 1, wherein the step of processing the relevant data of the power grid by the data processing module in step S2 comprises:
s21, collecting original related data of the power grid from the detection module;
s22, carrying out noise processing on the original related data of the power grid;
s23, converting the related data of the power grid subjected to noise processing into a standard format so that the related data is in a universal range;
S24, resampling the related data of the power grid converted into the standard mode, ensuring uniformity of a time sequence, and dividing continuous data into windows with fixed sizes for local analysis and feature extraction.
3. The machine learning based power quality and voltage monitoring calibration method according to claim 1, characterized in that in step S3, the frequency component characteristics, harmonic distortion characteristics, phase angle characteristics, power factor characteristics, load fluctuation characteristics of the key nodes at the sampling time t are obtained for normalization, and a characteristic vector FeatureVector is composed:
;
wherein ω ' is normalized phase angle characteristic, PE ' is normalized power factor characteristic, load_f ' is normalized Load fluctuation characteristic, F ' is normalized frequency component characteristic, THD ' is normalized total harmonic distortion characteristic.
4. The machine learning based power quality and voltage monitoring calibration method of claim 1, wherein in step S6, the calibration module automatically calibrates the detection module after receiving the calibration request signal according to the following steps:
S61, after receiving the calibration request signal sent by the abnormality monitoring module, the calibration module compares the calculated state index ACT with a preset or historical reference value to determine whether calibration is needed;
S62, if the absolute value of the difference between the state index and the reference value is larger than a preset value, the calibration module sends a calibration signal to the detection module and jumps to S64, otherwise, the calibration is not needed, and the step S63 is jumped to;
s63, the detection module does not need to be adjusted, and the step S1 is returned;
S64, the detection module adjusts filter setting and sampling rate based on the calibration signal, and jumps to step S65;
S65, after adjustment is completed, the current moment is obtained again to serve as the sampling time t, the steps S1-S62 are repeatedly executed, the calibration effect is verified through the step S62, whether the current detection module still needs to be calibrated is judged, if the calibration effect meets the condition that the absolute value of the difference between the state index and the reference value is larger than a preset value, the step S64 is skipped, and if the calibration effect does not meet the condition that the absolute value of the difference between the state index and the reference value is larger than the preset value, the step S63 is skipped.
5. The machine learning based power quality and voltage monitoring calibration method of claim 3 wherein the normalized Load fluctuation signature load_f' is calculated according to the following equation:
;
wherein min (load_f) is the minimum value of the Load fluctuation feature in the history data of the ith key node, max (load_f) is the maximum value of the Load fluctuation feature in the history data of the ith key node, and load_f is the Load fluctuation feature, and the value is calculated according to the following formula:
;
Max (Load) and min (Load) are the maximum value and the minimum value of the Load in the observation period respectively, and Average (Load) is the Average value of the Load in the historical data of the key nodes of the power grid.
6. The power quality and voltage monitoring and calibrating system based on machine learning realizes the power quality and voltage monitoring and calibrating method based on machine learning according to any one of claims 1-5, wherein the power quality and voltage monitoring and calibrating system comprises a server, a detection module, a data processing module, a feature extraction module, a machine learning module, an abnormality monitoring module and a calibrating module, and the server stores intermediate data and process data of the detection module, the data processing module, the feature extraction module, the machine learning module, the abnormality monitoring module and the calibrating module;
The detection module acquires relevant data of key nodes on a power grid line, the data processing module is used for processing the relevant data so as to ensure the quality of the relevant data, the feature extraction module extracts key features from the processed relevant data, the machine learning module evaluates the key features to form an evaluation result, the abnormality monitoring module calculates a state index ACT of power quality and voltage monitoring according to the evaluation result and judges whether the state index ACT is abnormal, and the calibration module automatically calibrates the detection module;
Wherein the key features include frequency component features, harmonic distortion features, phase angle features, power factor features, and load ripple features.
7. The machine learning based power quality and voltage monitoring calibration system of claim 6, wherein the detection module comprises a voltage detection unit that collects voltage data of a critical node of a power grid, a current detection unit that collects current data of the critical node of the power grid, and a data store that stores the voltage data and the current data.
8. The machine learning based power quality and voltage monitoring calibration system of claim 6, wherein the calibration module includes a calibration unit for alerting a manager of anomalies in power quality and voltage monitoring and an alert unit for calibrating the detection module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410348644.3A CN117949884B (en) | 2024-03-26 | 2024-03-26 | Power quality and voltage monitoring and calibrating method and system based on machine learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410348644.3A CN117949884B (en) | 2024-03-26 | 2024-03-26 | Power quality and voltage monitoring and calibrating method and system based on machine learning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117949884A CN117949884A (en) | 2024-04-30 |
CN117949884B true CN117949884B (en) | 2024-06-11 |
Family
ID=90792367
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410348644.3A Active CN117949884B (en) | 2024-03-26 | 2024-03-26 | Power quality and voltage monitoring and calibrating method and system based on machine learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117949884B (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103323686A (en) * | 2012-03-19 | 2013-09-25 | 上海利思电气有限公司 | Smart power grid power quality analyzer |
CN117634892A (en) * | 2024-01-24 | 2024-03-01 | 国网山东省电力公司禹城市供电公司 | Power metering data security risk assessment method and system |
CN117641157A (en) * | 2023-11-30 | 2024-03-01 | 国网辽宁省电力有限公司营销服务中心 | Meter reading method of electricity consumption information acquisition terminal |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3745150B1 (en) * | 2019-05-27 | 2023-11-08 | Siemens Aktiengesellschaft | Method and device for determining the degradation of a battery module or battery cell |
-
2024
- 2024-03-26 CN CN202410348644.3A patent/CN117949884B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103323686A (en) * | 2012-03-19 | 2013-09-25 | 上海利思电气有限公司 | Smart power grid power quality analyzer |
CN117641157A (en) * | 2023-11-30 | 2024-03-01 | 国网辽宁省电力有限公司营销服务中心 | Meter reading method of electricity consumption information acquisition terminal |
CN117634892A (en) * | 2024-01-24 | 2024-03-01 | 国网山东省电力公司禹城市供电公司 | Power metering data security risk assessment method and system |
Also Published As
Publication number | Publication date |
---|---|
CN117949884A (en) | 2024-04-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20140156094A1 (en) | Sigma algebraic approximants as a diagnostic tool in power networks | |
EP3896776A1 (en) | Simulated battery construction method and simulated battery construction device | |
CN117233687B (en) | CVT initial error assessment method, medium and terminal based on historical data | |
CN114660531B (en) | Detection method, system and device based on ammeter measurement error compensation | |
CN114626769B (en) | Operation and maintenance method and system for capacitor voltage transformer | |
CN117949884B (en) | Power quality and voltage monitoring and calibrating method and system based on machine learning | |
CN117347916A (en) | Winding structure fault positioning method and device based on parameter identification | |
CN117195647A (en) | Method, apparatus, device, medium and program product for post-earthquake evaluation of transformer bushings | |
CN116840767A (en) | Electric energy metering device overall error assessment method and device, storage medium and terminal | |
CN114664392B (en) | Electrochemical parameter prediction method, device, electronic equipment and readable storage medium | |
CN107368000B (en) | A kind of Room Power environment control method | |
CN113125037B (en) | Cable conductor temperature estimation method based on distributed optical fiber on-line temperature measurement system | |
CN115267641A (en) | Method and system for identifying error abnormity of current transformer in same-tower double-circuit power transmission line | |
CN113746132B (en) | Photovoltaic power station based on cloud edge cooperation and control method thereof | |
CN115146715A (en) | Power utilization potential safety hazard diagnosis method, device, equipment and storage medium | |
CN114897006A (en) | Intelligent ammeter error classification method, system, equipment and storage medium | |
CN118091527B (en) | Voltage transformer error assessment method based on interpretive deep learning | |
CN117723917B (en) | Monitoring application method based on optical fiber extrinsic Fabry-Perot type ultrasonic sensor | |
CN117741321B (en) | Mobile storage and charging system fault diagnosis method and system | |
CN113702780B (en) | BP neural network-based high-voltage power supply online monitoring method and device | |
CN116500532B (en) | Metering abnormality evaluation method and system for synchronous cross sampling high-voltage potential transformer | |
CN115716217B (en) | Method and device for detecting spindle runout abnormality and storage medium | |
CN112103911B (en) | Hidden fault discrimination method and device for relay protection system | |
CN113504498B (en) | Performance detection method and system of partial discharge intelligent sensing terminal | |
CN116817983A (en) | Data analysis method, data analysis recorder and storage 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 |