CN117391481B - Big data-based power data monitoring method and system - Google Patents
Big data-based power data monitoring method and system Download PDFInfo
- Publication number
- CN117391481B CN117391481B CN202311687634.4A CN202311687634A CN117391481B CN 117391481 B CN117391481 B CN 117391481B CN 202311687634 A CN202311687634 A CN 202311687634A CN 117391481 B CN117391481 B CN 117391481B
- Authority
- CN
- China
- Prior art keywords
- data
- power
- curve segment
- curve
- abnormal
- 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
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000012544 monitoring process Methods 0.000 title claims abstract description 46
- 230000002159 abnormal effect Effects 0.000 claims abstract description 163
- 230000008859 change Effects 0.000 claims abstract description 67
- 238000012937 correction Methods 0.000 claims abstract description 36
- 238000012216 screening Methods 0.000 claims abstract description 7
- 230000005856 abnormality Effects 0.000 claims description 45
- 238000001514 detection method Methods 0.000 claims description 14
- 238000005265 energy consumption Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 8
- 230000035772 mutation Effects 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 description 18
- 238000004458 analytical method Methods 0.000 description 6
- 230000007704 transition Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 230000003247 decreasing effect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
- G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the field of power data monitoring, in particular to a power data monitoring method and system based on big data. The method comprises the steps of firstly obtaining a time sequence of power data contained in power equipment of the same type, segmenting a power change curve fitted through the time sequence to obtain curve segments, screening operation phases and standby phases of the power equipment from all the curve segments, detecting abnormal data of the power equipment from the operation phases and the standby phases respectively, correcting the corresponding abnormal data through the obtained abnormal degree to obtain a correction sequence of each power equipment, predicting future power data of the power equipment based on the correction sequence to obtain prediction data, and monitoring the power data of the power equipment in real time according to the prediction data. The method and the device can improve the accuracy of future power data prediction of the power equipment, and further improve the accuracy of real-time power data monitoring.
Description
Technical Field
The invention relates to the field of power data monitoring, in particular to a power data monitoring method and system based on big data.
Background
In the process of monitoring the energy consumption condition of the power equipment, the historical power data generated in the production process of the power equipment is collected, analyzed and compared with the real-time power data, so that the abnormality in the power data is found, and the real-time monitoring of the energy consumption condition of the power equipment is realized.
In the related art, future power data is usually predicted by analyzing historical power data, and real-time power data and a prediction result are compared and analyzed to achieve the effect of real-time monitoring, but because of the reasons of self faults of power equipment and the like, the historical power data is easy to be interfered by noise to form abnormal data, the accuracy of the predicted power data is lower, and therefore the accuracy of monitoring the power data is reduced.
Disclosure of Invention
In order to solve the technical problem that collected historical electric power data is easy to be interfered by noise to form abnormal data, and the accuracy of the predicted electric power data is low, so that the accuracy of electric power data monitoring is reduced, the invention aims to provide an electric power data monitoring method and system based on big data, and the adopted technical scheme is as follows:
the invention provides a power data monitoring method based on big data, which comprises the following steps:
acquiring a time sequence of the same type of power equipment in a preset first time period, wherein the time sequence comprises power data at different moments;
performing curve fitting on the time sequence of each power device to obtain a power change curve; segmenting the electric power change curve to obtain different curve segments; screening the operation stage and the standby stage of the power equipment from all curve segments according to the change of the power data on the curve segments; respectively carrying out anomaly detection on the electric power data in the operation stage and the standby stage to obtain anomaly data of the electric power equipment;
obtaining the abnormality degree of each abnormal data in the power equipment according to the difference between each abnormal data in the power equipment and other power data in the corresponding power equipment and the difference between the corresponding abnormal data and the power data of other power equipment at the corresponding time; correcting the corresponding abnormal data in the time sequence based on the abnormal degree to obtain a correction sequence of each power device;
predicting the power data in a preset second time period based on the correction sequence to obtain prediction data; and monitoring the power data in real time based on the prediction data.
Further, the screening the operation phase and the standby phase of the power equipment from all curve segments according to the change of the power data on the curve segments comprises:
taking the average value of the power data on each curve segment as the integral power of the corresponding curve segment;
taking the slope of a connecting line between the starting point and the end point of each curve segment as the power change degree of the corresponding curve segment;
the curve sections except the first curve section and the last curve section on the power change curve are used as target curve sections;
taking a previous curve segment adjacent to each target curve segment as a predecessor curve segment of the corresponding target curve segment, and taking a next curve segment adjacent to each target curve segment as a successor curve segment of the corresponding target curve segment;
obtaining phase distinguishing parameters of corresponding target curve segments according to the difference of the integral power between the subsequent curve segment and the predecessor curve segment corresponding to each target curve segment and the difference of the power change degree between the corresponding target curve segment and the subsequent curve segment;
taking the target curve segment corresponding to the maximum value of the phase distinguishing parameter as a starting phase of the power equipment, and taking the target curve segment corresponding to the minimum value of the phase distinguishing parameter as a stopping phase of the power equipment;
taking the starting stage, the stopping stage and a curve segment between the starting stage and the stopping stage as an operation stage of the power equipment; and taking other curve segments except the operation phase as standby phases of the power equipment.
Further, the step of obtaining the phase distinguishing parameter of the corresponding target curve segment according to the difference of the overall power between the subsequent curve segment and the predecessor curve segment corresponding to each target curve segment and the difference of the power variation degree between the corresponding target curve segment and the subsequent curve segment includes:
normalizing the difference value of the integral power of the subsequent curve segment and the integral power of the predecessor curve segment to obtain a data change parameter corresponding to the target curve segment;
normalizing the absolute value of the difference value between the power change degree of each target curve segment and the power change degree of the corresponding subsequent curve segment to obtain a degree change parameter of the corresponding target curve segment;
and taking the absolute value of the electric power change degree of each target curve segment, the product value of the data change parameter and the degree change parameter as a stage distinguishing parameter of the corresponding target curve segment.
Further, the performing anomaly detection on the power data in the operation stage and the standby stage respectively, and obtaining the anomaly data of the power equipment includes:
clustering the power data in the operation stage and the standby stage respectively to obtain clusters;
and taking all the power data outside the cluster as abnormal data of the corresponding power equipment.
Further, the obtaining the abnormality degree of each abnormal data in the power device according to the difference between each abnormal data in the power device and other power data in the corresponding power device and the difference between the corresponding abnormal data and the power data of the other power device at the corresponding time includes:
wherein,indicate->First->Degree of abnormality of the individual abnormal data; />Indicate->The>Abnormal data; />Indicate->Individual electric powerAn average of all power data in the device; />Indicate->First->Degree of mutation of the individual anomaly data; />Is indicated at +.>First->At the time of the abnormal data +.>Power data of the other power devices; />Representing the number of electrical devices; />Indicate->First->The power data of the moment before the moment of the abnormal data; />Indicate->First->Personal exception dataPower data at a time subsequent to the time at which the power data was received; />Representing natural constants; />Representing the normalization function.
Further, the correcting the corresponding abnormal data in the time series based on the abnormal degree, and obtaining the corrected series of each power device includes:
taking the difference value between the average value of all the power data of each power device and each abnormal data as the deviation value of the corresponding abnormal data;
taking the product value of the abnormality degree and the deviation value of each abnormal data as an adjustment value of the corresponding abnormal data;
taking the sum value of each abnormal data and the corresponding adjustment value as correction data of the corresponding abnormal data;
and replacing all abnormal data in the time sequence with corresponding correction data to obtain a correction sequence of each power device.
Further, the real-time monitoring of the power data based on the prediction data includes:
if the real-time power data of the power equipment are different from the predicted data, the energy consumption of the corresponding power equipment is abnormal; otherwise, the energy consumption of the corresponding power equipment is normal.
Further, the segmenting the power variation curve to obtain different curve segments includes:
obtaining extreme points in the power change curve based on Newton's method;
and segmenting the power change curve by taking the extreme point as a segmentation point to obtain different curve segments.
Further, the power data prediction method is an autoregressive integral moving average model.
The invention also provides a big data-based power data monitoring system, which comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes any one step of the big data-based power data monitoring method when executing the computer program.
The invention has the following beneficial effects:
the invention considers that the historical power data is easy to be interfered by factors such as noise and the like to have abnormal data, and reduces the accuracy of a prediction result, thereby reducing the accuracy of monitoring the power data, and therefore, the abnormal data needs to be corrected. In consideration of the large difference of the power data of the power equipment in the running state and the standby state, the abnormal data is directly extracted from all the data, the accuracy of detecting the abnormal data is reduced, the change trend of the power data in different states is different, and the power data is obviously increased or reduced when the power data is transited from one state to the other state, so that the power change curve is firstly segmented, the change trend of the power data of each curve segment is analyzed, the running stage and the standby stage are distinguished, the abnormal data is detected in the data in the same stage, and the accuracy of detecting the abnormal data in different states of the power equipment is improved; and further, the corresponding abnormal data is corrected to different degrees according to the obtained abnormal degree, so that the correction precision of the abnormal data is improved, the corrected data is more similar to real data, the accuracy of the subsequent future data prediction is improved, and the accuracy of the power data monitoring is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a power data monitoring method based on big data according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description refers to the specific implementation, structure, characteristics and effects of the method and the system for monitoring electric power data based on big data according to the invention by combining the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
An embodiment of a power data monitoring method and system based on big data:
the following specifically describes a specific scheme of the power data monitoring method and system based on big data provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for monitoring power data based on big data according to an embodiment of the present invention is shown, where the method includes:
step S1: and acquiring a time sequence of the same type of power equipment in a preset first time period, wherein the time sequence comprises power data at different moments.
In the process of monitoring the energy consumption condition of the power equipment, the historical power data is usually utilized to predict future power data, and real-time power data and a prediction result are compared and analyzed to achieve the effect of real-time monitoring, but due to the fact that the power equipment is high in internal temperature and the like caused by faults or long working time, the historical power data is easy to be interfered by noise to form abnormal data, the accuracy of the predicted power data is low, and therefore the accuracy of monitoring the power data and monitoring the energy consumption condition of the power equipment is reduced. The embodiment of the invention provides a power data monitoring method based on big data to solve the problem.
In the embodiment of the invention, a special power monitoring instrument such as a current monitor or a voltage detector is firstly used for collecting power data of a plurality of power devices of the same type in a preset first time period, and a sequence formed by the power data at different moments is used as a time sequence corresponding to the power devices, wherein the collected power data in one embodiment of the invention is the current of the power devices, and the voltage or the electric power of the power devices can be collected as the power data in other embodiments of the invention, and the invention is not limited herein. In one embodiment of the present invention, the preset first period is set to 1 day, and the power data is collected every 1 minute, and specific values of the preset first period and the time interval for collecting the power data may be set by the practitioner according to the specific implementation scenario, which is not limited herein.
After the time sequence of each power device containing the power data is obtained, the time sequence of each power device can be combined for analysis in the follow-up, and abnormal data in the time sequence can be detected and corrected.
Step S2: performing curve fitting on the time sequence of each power device to obtain a power change curve; segmenting the electric power change curve to obtain different curve segments; screening the operation stage and the standby stage of the power equipment from all curve segments according to the change of the power data on the curve segments; and respectively carrying out anomaly detection on the power data in the operation stage and the standby stage to obtain the anomaly data of the power equipment.
The power equipment is subject to noise interference due to self faults or overlong working time, excessively high internal temperature and other reasons, so that abnormal data exist, and therefore, in order to improve the accuracy of predicting future power data according to the collected power data, the abnormal data in the power equipment need to be detected and corrected. Since the main states of the power equipment are an operation state and a standby state, the operation state is an active state in actual operation, the standby state is a state for keeping certain functions such as a standby indicator light on, and the difference of power data in the two states is large, the power data in the standby state is generally smaller than the power data in the operation state, if abnormal data is directly detected in all the power data of a certain power equipment, the normal and large power data in the operation state is mistakenly regarded as the abnormal data, or the normal and small power data in the standby state is mistakenly regarded as the abnormal data, so that the abnormal detection needs to be carried out on the power data in the operation state and the standby state respectively, and the accuracy of the abnormal data detection of the power equipment is improved.
In order to reduce the influence of sudden increase or sudden decrease of current on a circuit and equipment in the starting and closing process of the power equipment, a soft start and soft stop technology is generally adopted, and the starting and stopping process of the power equipment is smoothed by gradually increasing or gradually decreasing the current, so that the embodiment of the invention firstly carries out curve fitting on a time sequence of each power equipment to obtain a corresponding power change curve, then segments the power change curve to obtain a plurality of curve sections therein, and the analysis of power data on the curve sections is facilitated in the follow-up process, and the operation stage and the standby stage in the curve sections are distinguished. In one embodiment of the present invention, the time series sequence is subjected to curve fitting by using a polynomial curve fitting method, and other curve fitting methods may be used in other embodiments of the present invention, which are not limited herein, and polynomial curve fitting is a technical means well known to those skilled in the art, and will not be described herein.
Preferably, in one embodiment of the present invention, the method for acquiring different curve segments in the power variation curve specifically includes:
because the power data can be increased or decreased in the process of the transition of the power equipment from one state to the other state, the extreme point in the power change curve can be obtained based on the Newton method; the extreme points are used as dividing points to divide the power change curve to obtain different curve segments, wherein each curve segment can be an operation stage of the power equipment or a standby stage of the power equipment, so that further analysis on power data on the curve segment is needed in the follow-up process, and the operation stage and the standby stage of the power equipment are distinguished.
According to the analysis, the power equipment has larger power data in the running state and smaller power data in the standby state, and the power data can gradually increase or decrease in the transition process from one state to the other state, so that the change condition of the power data on the curve segments can be analyzed, and the running stage and the standby stage of the power equipment are screened from all the curve segments, so that the accuracy of the subsequent abnormal detection of the power data is ensured.
Preferably, in one embodiment of the present invention, the method for acquiring the operation phase and the standby phase of the power device specifically includes:
taking the average value of the power data on each curve segment as the integral power of the corresponding curve segment; taking the slope of a connecting line between the starting point and the end point of each curve segment as the power change degree of the corresponding curve segment; the curve sections except the first curve section and the last curve section on the power change curve are used as target curve sections; taking a previous curve segment adjacent to each target curve segment as a predecessor curve segment of the corresponding target curve segment, and taking a next curve segment adjacent to each target curve segment as a successor curve segment of the corresponding target curve segment; obtaining phase distinguishing parameters of the corresponding target curve segments according to the difference of the integral power between the subsequent curve segment and the predecessor curve segment corresponding to each target curve segment and the difference of the power change degree between the corresponding target curve segment and the subsequent curve segment; taking a target curve segment corresponding to the maximum value of the phase distinguishing parameter as a starting phase of the power equipment, and taking a target curve segment corresponding to the minimum value of the phase distinguishing parameter as a stopping phase of the power equipment; the start phase and the stop phase and the curve segment between the two phases can be used as the operation phase of the power equipment; and taking other curve segments except the operation phase as standby phases of the power equipment.
Preferably, in one embodiment of the present invention, the method for acquiring the phase distinction parameter specifically includes:
normalizing the difference value of the integral power of the subsequent curve segment and the integral power of the predecessor curve segment to obtain a data change parameter corresponding to the target curve segment; normalizing the absolute value of the difference between the power change degree of each target curve segment and the power change degree of the corresponding subsequent curve segment to obtain a degree change parameter of the corresponding target curve segment; and taking the absolute value of the electric power change degree of each target curve segment, the product value of the data change parameter and the degree change parameter as the stage distinguishing parameter of the corresponding target curve segment. The expression of the phase discrimination parameter may specifically be, for example:
wherein,indicate->Stage distinguishing parameters of the target curve segments; />Indicate->The degree of power variation of the individual target curve segments; />Indicate->The degree of power variation of the subsequent curve segment of the individual target curve segment; />A maximum value representing the degree of power variation for all curve segments; />Indicate->Integral power of successive curve segments of the target curve segment;/>Indicate->Overall power of the predecessor curve segments of the individual target curve segments; />Representing the maximum of the overall power for all curve segments.
In the process of acquiring the phase distinguishing parameters of each target curve segment, the phase distinguishing parametersThe method has the function of identifying the starting stage and the stopping stage of the power equipment in all curve segments, so that the operation stage and the standby stage of the power equipment can be distinguished; since the start phase of the power equipment is a transition phase from the standby phase to the operation phase and the stop phase is a transition phase from the operation phase to the standby phase, the analysis shows that the power data shows a trend of gradually increasing or gradually decreasing when the power equipment is in soft start or soft stop, so that the absolute value of the power variation degree of the target curve section is->The larger the target curve segment, the more likely the target curve segment is a start-up phase or a stop phase of the power device; when the power equipment enters the operation stage or the standby stage, the change of the power data is stable, so that the degree of the target curve segment changes the parametersThe larger the target curve segment with larger variation trend enters a relatively stable curve segment, and further the target curve segment is more likely to be a starting stage or a stopping stage, wherein the maximum value of the electric power variation degree +.>For aligningNormalizing; since the power data of the power plant in the run phase is large and the power data in the standby phase is small, the parameters +.>Further confirming whether the target curve segment is a start-up phase or a stop phase, wherein the maximum value of the overall power +.>For->Normalizing, when the data change parameter is bigger and positive, the target curve segment is a transition phase from a standby phase to an operation phase, and further the target curve segment is more likely to be a starting phase, and when the data change parameter is smaller and negative, the target curve segment is more likely to be a stopping phase, because>、/>And->Is used as the phase differentiating parameter of the target curve segment +.>Whereas the phase discrimination parameter of the target curve segment may be smaller than 0 or larger than 0, so the phase discrimination parameter +.>The larger the corresponding target curve segment is, the more likely it is to be the start-up phase, the phase discrimination parameter +.>The smaller the indication the more likely the corresponding target curve segment is a stop phase.
After the operation phase and the standby phase of the power equipment are acquired, the power data in the operation phase and the standby phase can be detected abnormally, and the abnormal data of the power equipment can be identified.
Preferably, in one embodiment of the present invention, the method for acquiring abnormal data of the electrical device specifically includes:
the clustering algorithm can obtain outliers in the sample points, wherein the outliers can be regarded as abnormal points in the sample points, so that the power data in the operation stage and the standby stage can be clustered respectively to obtain clusters; and taking all the power data outside the cluster as abnormal data of the corresponding power equipment. In one embodiment of the present invention, the power data is clustered using a DBSCAN clustering algorithm, and in other embodiments of the present invention, the power data may be clustered using a K-means clustering algorithm, which is not limited herein.
In other embodiments of the present invention, the anomaly detection may be performed on the power data by using a box map or an isolated forest, which are all technical means known to those skilled in the art, and are not described herein.
After the abnormal data of each power device is obtained, the abnormal data can be corrected in the follow-up process, so that the accuracy of predicting the future power data can be improved.
Step S3: obtaining the abnormality degree of each abnormal data in the power equipment according to the difference between each abnormal data in the power equipment and other power data in the corresponding power equipment and the difference between the corresponding abnormal data and the power data of other power equipment at the corresponding time; and correcting the corresponding abnormal data in the time sequence based on the degree of abnormality to obtain a correction sequence of each power device.
Since the abnormal data in each power device needs to be corrected in the follow-up process, the abnormal degree of different abnormal data is different, the difference degree of the abnormal data can be reflected through the difference between the abnormal data and other power data in the power device, the greater the difference between the abnormal data and other power data of the power device is, the greater the abnormal degree of the abnormal data is indicated, meanwhile, the difference between the abnormal data and the power data of other power devices at the same moment can be reflected, the greater the difference between the abnormal data and the power data of other power devices at the same moment is indicated, the greater the abnormal degree of the abnormal data is indicated, the corresponding abnormal data can be corrected according to the abnormal degree in the follow-up process, and the accuracy of data correction is improved.
Preferably, in one embodiment of the present invention, the method for acquiring the abnormality degree of each abnormality data in the power equipment specifically includes:
firstly, acquiring an average value of all power data in each power device, taking an absolute value of a difference value between abnormal data and the average value of the power data as a deviation degree of the abnormal data, reflecting the deviation degree of the abnormal data from the whole power data by using the deviation degree, and indicating that the greater the deviation degree is, the greater the deviation degree of the abnormal data from the whole power data is, and the greater the abnormality degree of the abnormal data is; adding the absolute value of the difference value between the abnormal data and the power data at the previous and the next time respectively, wherein the larger the mutation degree of the abnormal data is, the larger the local change of the abnormal data is, and the larger the abnormality degree of the abnormal data is; further, the absolute value of the difference value of the abnormal data and the power data of other equipment at the same moment is taken as the difference degree of the abnormal data, the larger the difference degree is, the larger the difference between the abnormal data and the power data of other power equipment is, and the abnormal degree of the abnormal data is obtained by combining the analysis of the average value of the difference degree, the deviation degree and the mutation degree of the abnormal data between the abnormal data and all other power equipment. The expression of the degree of abnormality may specifically be, for example:
wherein,indicate->First->Degree of abnormality of the individual abnormal data; />Indicate->The>Abnormal data; />Indicate->An average of all power data in the individual power devices; />Indicate->First->Degree of mutation of the individual anomaly data; />Is indicated at +.>First->At the time of the abnormal data +.>Power data of the other power devices; />Representing the number of electrical devices; />Indicate->First->The power data of the moment before the moment of the abnormal data; />Indicate->First->Power data at a time subsequent to the time at which the individual abnormal data is located; />Representing natural constants; />Representing the normalization function.
In the acquisition process of the degree of abnormality of each abnormality data in the power equipment,represents the degree of deviation of each abnormal data, degree of deviation +>The larger the abnormality data is, the more the abnormality data is deviated from the power data of the power equipment as a whole, the degree of abnormality of the abnormality data is +>The larger, therein->Indicating the degree of deviationNormalizing; />The degree of mutation of each abnormal data with respect to the power data at the previous and subsequent time points is shown +.>The larger the abnormality data is, the more the abnormality data deviates from the power data of the power equipment locally, the degree of abnormality of the abnormality data is +>The larger, therein->Indicates the degree of mutation->Normalizing; />Representing the degree of difference between the abnormal data and the power data of other devices at the same timeThe larger the difference between the abnormal data and the power data of other power equipment at the same time is, the degree of abnormality of the abnormal data is +>The larger, and the average value of the degree of difference between each abnormal data and the power data of all other power devices at the same time is combined +.>Anomalies reflecting the anomalous dataThe product value of the three is further normalized to obtain the abnormality degree of abnormal data>Thereby limiting the degree of abnormality to +.>And in the range, the subsequent evaluation and adjustment of the abnormal data are convenient.
In one embodiment of the present invention, the normalization process may specifically be, for example, maximum and minimum normalization processes, and the normalization in the subsequent steps may be performed by using the maximum and minimum normalization processes, and in other embodiments of the present invention, other normalization methods may be selected according to a specific range of values, which will not be described herein.
After the abnormality degree of each abnormal data in each device is obtained, the corresponding abnormal data in the time sequence can be corrected based on the abnormality degree, so that a correction sequence of each power device is obtained, future power data of the power device can be predicted conveniently according to the correction sequence, and the accuracy of power data prediction is improved.
Preferably, in one embodiment of the present invention, the method for acquiring the correction sequence of each power device specifically includes:
taking the difference value between the average value of all the power data of each power device and each abnormal data as the deviation value of the corresponding abnormal data; taking the product value of the abnormality degree and the deviation value of each abnormal data as the adjustment value of the corresponding abnormal data; taking the sum value of each abnormal data and the corresponding adjustment value as correction data of the corresponding abnormal data; and replacing all abnormal data in the time sequence with corresponding correction data to obtain a correction sequence of each power device. The expression of the correction data may specifically be, for example:
wherein,indicate->First->Correction data of the abnormal data; />Indicate->The>Abnormal data; />Indicate->First->Degree of abnormality of the individual abnormal data; />Represent the firstAverage of all power data in the individual power devices.
In the process of acquiring the correction data corresponding to the abnormal data,representing the deviation value of the corresponding abnormal data due to the abnormal data +.>May be too large or too small, so that the average value of all the power data of the corresponding power equipment is differed from the abnormal data and the degree of deviation is utilized +.>Deviation value->Adjusting to make the abnormality degree as close to the average value of all the power data in the power equipment as possible after correction, taking the product of the abnormality degree and the average value as the final adjustment value of the abnormality data, and taking the sum of the abnormality data and the adjustment value as correction data->。
After the correction sequence of each power device is obtained, the power data of the corresponding power device in the future can be predicted based on the correction sequence, so that the accuracy of data prediction is improved, and the accuracy of monitoring the power data can be improved.
Step S4: predicting the power data in a preset second time period based on the correction sequence to obtain prediction data; the power data is monitored in real time based on the prediction data.
In order to realize real-time monitoring of power data and to find anomalies in power consumption of the power equipment, prediction data is needed to be obtained by predicting the power data in a preset second time period based on a correction sequence of the power equipment, so that the real-time monitoring of anomalies in the power data based on the prediction data is facilitated. In one embodiment of the present invention, the preset second time period is set to 1 day, and the specific value of the preset second time period may also be set by the practitioner according to the specific implementation scenario, which is not limited herein.
Because the energy consumption of the power equipment is in a direct proportion relation with the power data, after the predicted data of each power equipment in the preset second time period is obtained, the real-time power data of the power equipment can be monitored based on the preset data, so that whether the energy consumption of the power equipment is abnormal or not is judged.
Preferably, in one embodiment of the present invention, if the real-time power data of the power device is different from the predicted data, which indicates that the real-time power data of the power device is abnormal, the energy consumption of the corresponding power device is abnormal; otherwise, the energy consumption of the corresponding power equipment is normal.
One embodiment of the present invention provides a big data based power data monitoring system, the system comprising a memory, a processor and a computer program, wherein the memory is used for storing a corresponding computer program, the processor is used for running the corresponding computer program, and the computer program can realize the method described in steps S1 to S4 when running in the processor.
In summary, in the embodiment of the invention, firstly, a time sequence of electric power data of the same type in a preset first time period is obtained, curve fitting is performed on the time sequence to obtain an electric power change curve, the electric power change curve is segmented to obtain different curve segments, change characteristics of the electric power data on the curve segments are analyzed, operation phases and standby phases of the electric power equipment are screened out from all the curve segments, abnormal detection is performed on the electric power data in the operation phases and the standby phases respectively to obtain abnormal data of the electric power equipment, further, according to differences between each abnormal data in the electric power equipment and other electric power data corresponding to the electric power equipment and differences between each abnormal data in the electric power equipment and the electric power data corresponding to other electric power equipment, the abnormal degree of each abnormal data in the electric power equipment is obtained, correction sequences of the electric power equipment are obtained based on the abnormal degree, further, prediction is performed on the electric power data of a preset second time period based on the correction sequences to obtain prediction data, and real-time monitoring is performed on the electric power data of the electric power equipment according to the prediction data.
An embodiment of an anomaly detection and correction method for power data is provided:
in the prior art, abnormal data in data is detected by using a clustering method or an isolated forest method and the like, and the abnormal data is corrected. However, since the power data of the power equipment in the running state and the standby state have differences, the abnormality detection is directly performed from all the power data, so that the accuracy of detecting the abnormality data is reduced, and the accuracy of correcting the abnormality data is further reduced.
In order to solve the problem, the present embodiment provides an anomaly detection and correction method for power data, including:
step S1: and acquiring a time sequence of the same type of power equipment in a preset first time period, wherein the time sequence comprises power data at different moments.
Step S2: performing curve fitting on the time sequence of each power device to obtain a power change curve; segmenting the electric power change curve to obtain different curve segments; screening the operation stage and the standby stage of the power equipment from all curve segments according to the change of the power data on the curve segments; and respectively carrying out anomaly detection on the power data in the operation stage and the standby stage to obtain the anomaly data of the power equipment.
Step S3: obtaining the abnormality degree of each abnormal data in the power equipment according to the difference between each abnormal data in the power equipment and other power data in the corresponding power equipment and the difference between the corresponding abnormal data and the power data of other power equipment at the corresponding time; and correcting the corresponding abnormal data in the time sequence based on the degree of abnormality to obtain a correction sequence of each power device.
The details of the steps S1 to S3 in the embodiment of the method and the system for monitoring electric power data based on big data are given above, and are not described herein again.
The beneficial effects brought by the embodiment are as follows: the invention considers that the historical power data is easy to be interfered by factors such as noise and the like to cause abnormal data, so that the abnormal data needs to be corrected, and because the power data of the power equipment in an operation state and a standby state have larger difference, the abnormal data is directly extracted from all the data, the accuracy of detecting the abnormal data is reduced, the change trend of the power data in different states is different, and the power data is obviously increased or reduced when the power data transits from one state to the other state, the invention firstly segments a power change curve and analyzes the change trend of the power data of each curve segment, thereby distinguishing the operation stage and the standby stage, detecting the abnormal data in the same stage, and improving the accuracy of detecting the abnormal data in different states of the power equipment; and further, the corresponding abnormal data is corrected to different degrees according to the obtained abnormal degree, so that the correction precision of the abnormal data is improved, and the corrected data is more similar to real data.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (9)
1. A method of power data monitoring based on big data, the method comprising:
acquiring a time sequence of the same type of power equipment in a preset first time period, wherein the time sequence comprises power data at different moments;
performing curve fitting on the time sequence of each power device to obtain a power change curve; segmenting the electric power change curve to obtain different curve segments; screening the operation stage and the standby stage of the power equipment from all curve segments according to the change of the power data on the curve segments; respectively carrying out anomaly detection on the electric power data in the operation stage and the standby stage to obtain anomaly data of the electric power equipment;
obtaining the abnormality degree of each abnormal data in the power equipment according to the difference between each abnormal data in the power equipment and other power data in the corresponding power equipment and the difference between the corresponding abnormal data and the power data of other power equipment at the corresponding time; correcting the corresponding abnormal data in the time sequence based on the abnormal degree to obtain a correction sequence of each power device;
predicting the power data in a preset second time period based on the correction sequence to obtain prediction data; monitoring the power data in real time based on the prediction data;
according to the difference between each piece of abnormal data in the power equipment and other pieces of power data in the corresponding power equipment and the difference between the corresponding piece of abnormal data and the power data of the other power equipment at the corresponding time, obtaining the degree of abnormality of each piece of abnormal data in the power equipment comprises:
wherein,indicate->First->Degree of abnormality of the individual abnormal data; />Indicate->The>Abnormality ofData; />Indicate->An average of all power data in the individual power devices; />Indicate->First->Degree of mutation of the individual anomaly data; />Is indicated at +.>First->At the time of the abnormal data +.>Power data of the other power devices; />Representing the number of electrical devices; />Indicate->First->The power data of the moment before the moment of the abnormal data; />Indicate->First->Power data at a time subsequent to the time at which the individual abnormal data is located; />Representing natural constants; />Representing the normalization function.
2. A method of monitoring electrical data based on big data as claimed in claim 1, wherein said screening all of the curve segments for operational and standby phases of the electrical device based on the change in electrical data on the curve segments comprises:
taking the average value of the power data on each curve segment as the integral power of the corresponding curve segment;
taking the slope of a connecting line between the starting point and the end point of each curve segment as the power change degree of the corresponding curve segment;
the curve sections except the first curve section and the last curve section on the power change curve are used as target curve sections;
taking a previous curve segment adjacent to each target curve segment as a predecessor curve segment of the corresponding target curve segment, and taking a next curve segment adjacent to each target curve segment as a successor curve segment of the corresponding target curve segment;
obtaining phase distinguishing parameters of corresponding target curve segments according to the difference of the integral power between the subsequent curve segment and the predecessor curve segment corresponding to each target curve segment and the difference of the power change degree between the corresponding target curve segment and the subsequent curve segment;
taking the target curve segment corresponding to the maximum value of the phase distinguishing parameter as a starting phase of the power equipment, and taking the target curve segment corresponding to the minimum value of the phase distinguishing parameter as a stopping phase of the power equipment;
taking the starting stage, the stopping stage and a curve segment between the starting stage and the stopping stage as an operation stage of the power equipment; and taking other curve segments except the operation phase as standby phases of the power equipment.
3. A method of monitoring electrical power data based on big data as claimed in claim 2, wherein the obtaining the phase discrimination parameters of the corresponding target curve segment based on the difference in the overall electrical power between the subsequent curve segment and the preceding curve segment corresponding to each target curve segment and the difference in the degree of electrical power variation between the corresponding target curve segment and the subsequent curve segment comprises:
normalizing the difference value of the integral power of the subsequent curve segment and the integral power of the predecessor curve segment to obtain a data change parameter corresponding to the target curve segment;
normalizing the absolute value of the difference value between the power change degree of each target curve segment and the power change degree of the corresponding subsequent curve segment to obtain a degree change parameter of the corresponding target curve segment;
and taking the absolute value of the electric power change degree of each target curve segment, the product value of the data change parameter and the degree change parameter as a stage distinguishing parameter of the corresponding target curve segment.
4. The method for monitoring power data based on big data according to claim 1, wherein the performing anomaly detection on the power data in the operation phase and the standby phase, respectively, to obtain anomaly data of the power equipment comprises:
clustering the power data in the operation stage and the standby stage respectively to obtain clusters;
and taking all the power data outside the cluster as abnormal data of the corresponding power equipment.
5. The method for monitoring power data based on big data according to claim 1, wherein the correcting the corresponding abnormal data in the time series based on the degree of abnormality, obtaining a corrected series for each power device, comprises:
taking the difference value between the average value of all the power data of each power device and each abnormal data as the deviation value of the corresponding abnormal data;
taking the product value of the abnormality degree and the deviation value of each abnormal data as an adjustment value of the corresponding abnormal data;
taking the sum value of each abnormal data and the corresponding adjustment value as correction data of the corresponding abnormal data;
and replacing all abnormal data in the time sequence with corresponding correction data to obtain a correction sequence of each power device.
6. The method of claim 1, wherein the real-time monitoring of the power data based on the predicted data comprises:
if the real-time power data of the power equipment are different from the predicted data, the energy consumption of the corresponding power equipment is abnormal; otherwise, the energy consumption of the corresponding power equipment is normal.
7. The method of claim 1, wherein segmenting the power profile to obtain different profile segments comprises:
obtaining extreme points in the power change curve based on Newton's method;
and segmenting the power change curve by taking the extreme point as a segmentation point to obtain different curve segments.
8. The method of claim 1, wherein the method of power data prediction is an autoregressive integral moving average model.
9. A big data based power data monitoring system, the system comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1-8 when the computer program is executed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311687634.4A CN117391481B (en) | 2023-12-11 | 2023-12-11 | Big data-based power data monitoring method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311687634.4A CN117391481B (en) | 2023-12-11 | 2023-12-11 | Big data-based power data monitoring method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117391481A CN117391481A (en) | 2024-01-12 |
CN117391481B true CN117391481B (en) | 2024-03-01 |
Family
ID=89466903
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311687634.4A Active CN117391481B (en) | 2023-12-11 | 2023-12-11 | Big data-based power data monitoring method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117391481B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117800187B (en) * | 2024-02-29 | 2024-05-10 | 凯尔菱电(山东)电梯有限公司 | Intelligent monitoring method and system for safe running state of elevator |
CN117828408B (en) * | 2024-03-04 | 2024-05-14 | 新风光电子科技股份有限公司 | Energy storage capacity data processing method and system |
CN118070683B (en) * | 2024-04-22 | 2024-07-23 | 山东师范大学 | Logistics transportation big data acquisition method and system based on digital twinning |
CN118137675B (en) * | 2024-04-30 | 2024-09-10 | 江苏天南电力股份有限公司 | Energy consumption balancing method and system for transmission line sensor |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104766175A (en) * | 2015-04-16 | 2015-07-08 | 东南大学 | Power system abnormal data identifying and correcting method based on time series analysis |
CN110032670A (en) * | 2019-04-17 | 2019-07-19 | 腾讯科技(深圳)有限公司 | Method for detecting abnormality, device, equipment and the storage medium of time series data |
CN111695620A (en) * | 2020-06-08 | 2020-09-22 | 中国电力科学研究院有限公司 | Method and system for detecting and correcting abnormal data of time sequence of power system |
CN112231361A (en) * | 2020-10-28 | 2021-01-15 | 西安热工研究院有限公司 | Wind power project generating capacity evaluation method based on fan operation data |
WO2023000482A1 (en) * | 2021-07-23 | 2023-01-26 | 广州新科佳都科技有限公司 | Platform door abnormality detection method and apparatus based on mechanism analysis |
CN116071521A (en) * | 2023-02-20 | 2023-05-05 | 东方电气集团东方电机有限公司 | Method, device, equipment and storage medium for drawing test curve of water turbine model |
CN117057517A (en) * | 2023-10-12 | 2023-11-14 | 国网吉林省电力有限公司长春供电公司 | Efficient processing method and system for electric power data based on digital twin |
-
2023
- 2023-12-11 CN CN202311687634.4A patent/CN117391481B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104766175A (en) * | 2015-04-16 | 2015-07-08 | 东南大学 | Power system abnormal data identifying and correcting method based on time series analysis |
CN110032670A (en) * | 2019-04-17 | 2019-07-19 | 腾讯科技(深圳)有限公司 | Method for detecting abnormality, device, equipment and the storage medium of time series data |
CN111695620A (en) * | 2020-06-08 | 2020-09-22 | 中国电力科学研究院有限公司 | Method and system for detecting and correcting abnormal data of time sequence of power system |
CN112231361A (en) * | 2020-10-28 | 2021-01-15 | 西安热工研究院有限公司 | Wind power project generating capacity evaluation method based on fan operation data |
WO2023000482A1 (en) * | 2021-07-23 | 2023-01-26 | 广州新科佳都科技有限公司 | Platform door abnormality detection method and apparatus based on mechanism analysis |
CN116071521A (en) * | 2023-02-20 | 2023-05-05 | 东方电气集团东方电机有限公司 | Method, device, equipment and storage medium for drawing test curve of water turbine model |
CN117057517A (en) * | 2023-10-12 | 2023-11-14 | 国网吉林省电力有限公司长春供电公司 | Efficient processing method and system for electric power data based on digital twin |
Non-Patent Citations (1)
Title |
---|
基于时间序列的配电网数据清理和融合方法研究;朱有产 等;《电网技术》;20210731;第45卷(第7期);2839-2846 * |
Also Published As
Publication number | Publication date |
---|---|
CN117391481A (en) | 2024-01-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN117391481B (en) | Big data-based power data monitoring method and system | |
CN112527788B (en) | Method and device for detecting and cleaning abnormal value of transformer monitoring data | |
CN109387712B (en) | Non-invasive load detection and decomposition method based on state matrix decision tree | |
CN109740648B (en) | Method and device for identifying abnormal data of power load and computer equipment | |
CN117312997B (en) | Intelligent diagnosis method and system for power management system | |
CN116955895B (en) | Transformer operation abnormity monitoring method and system | |
CN117193164B (en) | Fault monitoring method and system of numerical control machine tool | |
CN116434372B (en) | Intelligent data acquisition system and working condition identification system for variable working condition equipment | |
CN117411189A (en) | Monitoring data enhancement method of micro-grid coordination controller | |
CN117114213B (en) | Rural network co-construction convenience network service method and system | |
CN117290802A (en) | Host power supply operation monitoring method based on data processing | |
CN115828170B (en) | Fault detection method based on electronic control data of tunneling machine | |
CN117972618B (en) | Method and system for detecting secondary power failure of hybrid integrated circuit | |
CN117633695B (en) | Transformer operation monitoring method based on electrical parameter time sequence analysis | |
CN118378199A (en) | Real-time anomaly detection method in big data analysis platform | |
CN117331790A (en) | Machine room fault detection method and device for data center | |
CN114859270A (en) | Transformer winding fault diagnosis method and system | |
Bórnea et al. | Classification of bearing faults in induction motors with the hilbert-huang transform and feature selection | |
US20180087489A1 (en) | Method for windmill farm monitoring | |
Yu et al. | Novelty class detection in machine learning-based condition diagnosis | |
CN118311362B (en) | Running state monitoring method for energy-saving medium-power direct-current speed regulating device | |
CN118114082B (en) | Centrifuge data remote monitoring and analyzing system based on cloud computing | |
Lang et al. | Analysis of hydropower plant guide bearing vibrations by machine learning based identification of steady operations | |
CN117951633B (en) | Photovoltaic power generation equipment fault diagnosis method and system | |
CN118759298B (en) | Multi-parameter monitoring analysis method and device for transformer and transformer |
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 |