CN113326296B - Load decomposition method and system suitable for industrial and commercial users - Google Patents
Load decomposition method and system suitable for industrial and commercial users Download PDFInfo
- Publication number
- CN113326296B CN113326296B CN202110213501.8A CN202110213501A CN113326296B CN 113326296 B CN113326296 B CN 113326296B CN 202110213501 A CN202110213501 A CN 202110213501A CN 113326296 B CN113326296 B CN 113326296B
- Authority
- CN
- China
- Prior art keywords
- time
- characteristic
- difference
- determining
- feature
- 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 93
- 238000000354 decomposition reaction Methods 0.000 title claims abstract description 27
- 238000013507 mapping Methods 0.000 claims abstract description 9
- 230000001052 transient effect Effects 0.000 claims description 62
- 230000008569 process Effects 0.000 claims description 51
- 238000004364 calculation method Methods 0.000 claims description 37
- 230000005611 electricity Effects 0.000 claims description 26
- 230000008859 change Effects 0.000 claims description 23
- 238000007781 pre-processing Methods 0.000 claims description 18
- 238000001914 filtration Methods 0.000 claims description 10
- 238000010845 search algorithm Methods 0.000 claims description 9
- 230000035772 mutation Effects 0.000 claims description 8
- 230000010354 integration Effects 0.000 claims description 7
- 238000005191 phase separation Methods 0.000 claims description 4
- KJONHKAYOJNZEC-UHFFFAOYSA-N nitrazepam Chemical compound C12=CC([N+](=O)[O-])=CC=C2NC(=O)CN=C1C1=CC=CC=C1 KJONHKAYOJNZEC-UHFFFAOYSA-N 0.000 claims description 2
- 230000006870 function Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000002156 mixing Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008521 reorganization Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- 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
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Probability & Statistics with Applications (AREA)
- Fuzzy Systems (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Water Supply & Treatment (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Public Health (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a load decomposition method and a load decomposition system suitable for industrial and commercial users. Wherein the method comprises the following steps: extracting characteristics of waveform data according to a fixed time period to form a characteristic time sequence; based on DBSCAN clustering, clustering the preprocessed feature time sequence to obtain a clustering result, extracting sequence points of the preprocessed feature time sequence according to the clustering result, reconstructing a feature difference time sequence, and determining an optimized feature difference time sequence; performing difference matching on the optimized characteristic difference time sequence, and determining a difference matching result; mapping start-stop state events belonging to the same equipment, optimizing feature dimensions, identifying the electrical type of the equipment, determining a load identification result, and establishing a proprietary model feature library of industrial and commercial users; and for bus side waveform data to be analyzed for a period of time, according to a load identification result, obtaining the duty ratio of load power consumption of various devices for a period of time, and realizing the decomposition of the bus load of industrial and commercial users.
Description
Technical Field
The application relates to the technical field of power systems, in particular to a method and a system for decomposing loads of industrial and commercial users.
Background
The industrial and commercial users are taken as important components of the power users, the power load characteristics of the industrial and commercial users are in development trend of high duty ratio and diversity, and the industrial and commercial users have important influence on the upgrading, operation and maintenance of power grid sources, networks, loads and storages. At present, based on a non-intervention type load intelligent sensing technology, the working state, power and other electricity consumption information of various electric equipment in a user are identified and monitored by utilizing bus side electric information at a user metering point, so that more technical means are provided for various requirements such as environment protection monitoring, energy efficiency service, abnormal source positioning and the like. For industrial and commercial users, because of the numerous electric equipment related to different industries and the complex electric characteristics, a complete equipment characteristic library is difficult to build, and the load decomposition of different industries is difficult to realize.
Aiming at the technical problems that in the prior art, for industrial and commercial users, as electric equipment related to different industries is numerous, the electric characteristics are complex, a complete equipment characteristic library is difficult to build, and the load decomposition of different industries is difficult to realize, no effective solution is proposed at present.
Disclosure of Invention
The embodiment of the disclosure provides a method and a system suitable for load decomposition of industrial and commercial users, which at least solve the technical problems that in the prior art, as electric equipment related to different industries is numerous, electricity utilization characteristics are complex, a complete equipment characteristic library is difficult to build, and load decomposition of different industries is difficult to realize.
According to one aspect of an embodiment of the present disclosure, there is provided a method for load splitting suitable for business users, comprising: extracting characteristics of waveform data according to a fixed time period to form a characteristic time sequence, wherein the waveform data is data acquired by a bus side at high frequency; performing data preprocessing on the characteristic time sequence, and determining the preprocessed characteristic time sequence; based on DBSCAN clustering, clustering the preprocessed feature time sequence to obtain a clustering result, extracting sequence points of the preprocessed feature time sequence according to the clustering result, reconstructing a feature difference time sequence, and determining an optimized feature difference time sequence; performing difference matching on the optimized characteristic difference time sequence based on an optimal search algorithm, and determining a difference matching result; mapping start-stop state events belonging to the same equipment based on a difference matching result, optimizing feature dimensions, identifying the electrical type of the equipment, determining a load identification result, and establishing an exclusive model feature library of industrial and commercial users; and for bus side waveform data to be analyzed for a period of time, according to the load identification result, obtaining the duty ratio of the load power consumption of various devices for the period of time, and realizing the decomposition of the bus load of industrial and commercial users.
According to another aspect of the disclosed embodiments, there is also provided a system for load splitting for business users, comprising: the characteristic time sequence forming module is used for extracting characteristics of waveform data according to a fixed time period to form a characteristic time sequence, wherein the waveform data is data acquired by a bus side at high frequency; the characteristic time sequence determining module is used for carrying out data preprocessing on the characteristic time sequence and determining the characteristic time sequence after preprocessing; the module for determining the optimized characteristic difference time sequence is used for carrying out clustering operation on the preprocessed characteristic time sequence based on DBSCAN clustering to obtain a clustering result, extracting sequence points of the preprocessed characteristic time sequence according to the clustering result, reconstructing the characteristic difference time sequence, and determining the optimized characteristic difference time sequence; the difference value matching result determining module is used for performing difference value matching on the optimized characteristic difference value time sequence based on an optimal search algorithm, and determining a difference value matching result; the load identification result determining module is used for mapping start-stop state events belonging to the same equipment based on the difference value matching result, optimizing feature dimensions, identifying the electrical type of the equipment, determining a load identification result and establishing an exclusive model feature library of industrial and commercial users; and the device load power consumption ratio module is used for obtaining the power consumption ratio of various devices in a period of time according to the load identification result for bus side waveform data in a period of time to be analyzed, and realizing the decomposition of the industrial and commercial user bus load.
The invention provides a load decomposition method suitable for industrial and commercial users, which is an unsupervised load decomposition method, can complete the load decomposition of specific users without carrying out a large amount of data calibration and training, obtains typical electrical characteristics and equipment types of electric equipment of the industrial and commercial users, and has stronger robustness and self-adaption. Therefore, the method can be used for automatically adjusting parameters of the start-stop state algorithm of the specific user equipment in real time, the effectiveness of the algorithm is improved, and the corresponding labor cost can be effectively reduced.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate and explain the present disclosure, and together with the description serve to explain the present disclosure. In the drawings:
FIG. 1 is a schematic diagram of a method of load splitting suitable for business users according to an embodiment of the present disclosure;
FIG. 2 is a flow chart of an unsupervised load decomposition method according to an embodiment of the present disclosure;
FIG. 3 is a flow chart of a cluster-based feature difference sequence reorganization process according to an embodiment of the present disclosure;
FIG. 4 is a graph of raw load for operation of a plant mixing device according to an embodiment of the present disclosure;
FIG. 5 is an exploded stack-up view of the operation of certain plant mixing equipment according to an embodiment of the present disclosure;
FIG. 6 is a schematic diagram of a system suitable for load splitting for business users according to an embodiment of the present disclosure.
Detailed Description
The exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, however, the present invention may be embodied in many different forms and is not limited to the examples described herein, which are provided to fully and completely disclose the present invention and fully convey the scope of the invention to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. In the drawings, like elements/components are referred to by like reference numerals.
Unless otherwise indicated, terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art. In addition, it will be understood that terms defined in commonly used dictionaries should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense.
According to a first aspect of the present embodiment, a method 100 for load splitting for business users is provided. Referring to fig. 1, the method 100 includes:
s102, extracting characteristics of waveform data according to a fixed time period to form a characteristic time sequence, wherein the waveform data is data acquired by a bus side at high frequency;
s104, carrying out data preprocessing on the characteristic time sequence, and determining the preprocessed characteristic time sequence;
S106, clustering the preprocessed characteristic time sequence based on DBSCAN clustering to obtain a clustering result, extracting sequence points of the preprocessed characteristic time sequence according to the clustering result, reconstructing a characteristic difference time sequence, and determining an optimized characteristic difference time sequence;
s108, carrying out difference matching on the optimized characteristic difference time sequence based on an optimal search algorithm, and determining a difference matching result;
s110, mapping start-stop state events belonging to the same equipment based on a difference matching result, optimizing feature dimensions, identifying the electrical type of the equipment, determining a load identification result, and establishing an exclusive model feature library of industrial and commercial users;
And S112, for bus side waveform data to be analyzed for a period of time, according to the load identification result, obtaining the duty ratio of the load power consumption of various devices for the period of time, and realizing the decomposition of the bus load of industrial and commercial users.
The invention will be further elucidated with reference to the drawings and to specific embodiments.
The invention relates to a load decomposition method suitable for industrial and commercial users, which is an unsupervised load decomposition method, and the technical scheme of the invention is further described below with reference to the embodiment of specific application.
Referring to fig. 2, the method comprises the steps of:
step one, carrying out characteristic extraction of a fixed time period on waveform data acquired at high frequency at a bus side to form a characteristic time sequence;
Step two, carrying out data preprocessing operation on the characteristic time sequence;
Step three, as shown in fig. 3, extracting corresponding boundary points representing transient processes of equipment start-stop state change based on DBSCAN clustering, and reconstructing a characteristic difference time sequence;
step four, for the characteristic difference time sequence, completing the matching of the difference value based on an optimal search algorithm, wherein the matching corresponds to the matching of the start-stop state event of the same equipment;
Step five, further mapping the start-stop state events belonging to the same equipment based on a difference matching result, optimizing feature dimensions, identifying the electrical type of the equipment, and establishing a special model feature library of the user;
and step six, for bus side waveform data needing to be analyzed for a period of time, according to the load identification result of the step, obtaining the duty ratio of load power consumption of various devices in the period of time, and realizing the decomposition of the bus load of industrial and commercial users.
Further, in the first step:
setting a characteristic calculation time window for three-phase voltage and current waveforms acquired by the bus side at high frequency, carrying out split-phase characteristic calculation, and calculating AC two-phase load characteristics for high-count users; and for the high-supply low-supply users, calculating ABC three-phase load characteristics to form a characteristic time sequence. The following load characteristics are included in the sequence: three-phase current effective value, three-phase current 3 rd harmonic content (only for high-power low-power users), three-phase current 5 th harmonic content, three-phase current 7 th harmonic content, three-phase current 9 th harmonic content, three-phase current 11 th harmonic content, three-phase power factor and three-phase voltage distortion rate.
Further, the second step specifically comprises the following steps:
s2.1: the method comprises the steps of presetting a maximum threshold value of each feature dimension in a feature time sequence, carrying out outlier processing on the obtained feature time sequence, and when the feature exceeds the maximum threshold value, namely abnormal features appear, replacing the feature by using the corresponding feature in the previous unit time to eliminate the outlier.
S2.2: and presetting filter parameters, carrying out median filtering on the characteristic time sequence according to characteristic dimensions, and reserving mutation points to finish data preprocessing.
Further, the third step specifically comprises the following steps:
s3.1: performing DBSCAN clustering operation based on the three-phase current effective value/power time sequence, limiting the minimum neighborhood point number and the neighborhood radius, calibrating the time sequence point according to the clustering result, and only reserving a starting point and an ending point belonging to the same class in the time sequence;
S3.2: for the characteristic time sequence which only keeps the starting point and the ending point, calculating characteristic difference values among different categories according to time sequence in sequence to obtain a characteristic difference value time sequence, and simultaneously calculating a time difference value with category mutation as a characteristic quantity of newly added transient time in the sequence;
S3.3: presetting a device three-phase current effective value/power minimum judgment threshold value, sequentially combining partial differences in the characteristic difference time sequence according to judgment logic, and optimizing the characteristic difference time sequence.
Further, the fourth step specifically comprises the following steps:
S4.1: presetting a matching threshold value, and matching a difference value (representing the closing process of equipment) with negative three-phase current effective value/power in a characteristic difference value time sequence with a difference value sequence which is not matched before the moment according to a principle of preferentially matching a single difference value, a principle of preferentially matching the earliest moment and a principle of continuous difference value redundancy search based on an optimal search method, wherein if the difference value corresponding to the earliest moment in the difference value sequence to be matched is negative, deleting the difference value in the difference value sequence to be matched;
S4.2: and if the difference value combination is a combination type difference value match of non-single positive and negative difference values, the merging step is further completed for the difference values which are continuous in occurrence time and have the same attribute (both positive and negative).
Further, the fifth step specifically comprises the following steps:
s5.1: according to the difference value matching combination result in the step four, presetting an identity judgment threshold value of each characteristic dimension, classifying the matching result to obtain the variation mean value and the deviation coefficient of each characteristic quantity belonging to the same equipment in the start-stop process;
S5.2: according to the time mark corresponding to the starting transient process, extracting the corresponding characteristic time sequence before S2.2 filtering, calculating the impact parameter in the transient process as the newly added characteristic quantity, and the calculation formula is as follows:
In the above formula, M MAX and T MAX are the maximum values and corresponding time points of various features in the starting transient process, and M Steady and T Steady are various feature values and corresponding time points at the end of the transient process.
S5.3: based on univariate feature selection and a Pearson correlation coefficient method, weight ordering of feature quantities is obtained, feature selection is carried out according to sequence, features which can identify the most devices are preferentially selected until the selected feature combination can complete identification of all devices, and identifiers are marked on the selected features;
S5.4: and according to the actual use function requirement of the user, extracting comprehensive characteristics based on basic characteristics, marking identifiers on corresponding characteristics for the function application, and deleting useless characteristics by combining the characteristics for identification in S5.3.
Further, in the sixth step, calculating the power consumption ratio of each type of device specifically includes the following steps:
S6.1: the time nodes corresponding to the starting and stopping processes of various devices obtained according to the steps are divided into time windows, wherein the time nodes only comprise time nodes corresponding to the transient start in the starting transient process and time nodes corresponding to the transient end in the closing transient process;
s6.2: if the time window corresponds to a single device running, the power consumption of the corresponding device in the time window is the discrete integral of the three-phase active power value on the time window, and the calculation formula is as follows:
In the above formula, Δt is the characteristic calculation time period preset in the step one, p Ai、pBi、pCi is the corresponding three-phase active power value in the time window, and if the user is a high-count user, p Bi is 0.
Meanwhile, the average active power value of the equipment is obtained, and the calculation formula is as follows:
in the above formula, T M is the corresponding time point when the transient process ends, T N is the steady state time point farthest from the transient process, generally the starting time point or the ending time point of the time window, and if there is no obvious steady state process, the calculated time length is the whole time window length.
S6.3: if the time window corresponds to the operation of the combined equipment, the power consumption of the operation equipment (steady operation equipment) which does not generate state change in the time window is the discrete integration of the three-phase average active power value after the single operation of the equipment enters steady state in the time window, the average active power value is obtained by calculating the average value in other time windows corresponding to the state change of the equipment, and the calculation formula is as follows:
In the above And (3) calculating the average active power value of j-class equipment in each time window with state change, wherein T i is the corresponding steady-state calculation time.
Subtracting the sum of the average active power of various steady-state operation devices from the active power time sequence of the time window to obtain the active power time sequence of the single device with state change in the time window, wherein the calculation method of the power consumption and the average active power value of the device is the same as the formula (2) and the formula (3).
S6.3: the power consumption of each device in each time window in the period of time is classified and added to obtain the power consumption of each device in the period of time; and calculating the ratio of the power consumption of each device to the total power consumption in the period of time to obtain the power consumption ratio condition of each device. Referring to FIG. 4, a plot of the raw load for the operation of the plant mixing apparatus of FIG. 4 is shown. Referring to FIG. 5, FIG. 5 is an exploded stack-up view of the operation of a plant mixing apparatus.
Therefore, the invention provides a load decomposition method suitable for industrial and commercial users, which is an unsupervised load decomposition method, can complete the load decomposition of specific users without carrying out a large amount of data calibration and training, obtains typical electrical characteristics and equipment types of electric equipment of the industrial and commercial users, and has stronger robustness and self-adaption. Therefore, the method can be used for automatically adjusting parameters of the start-stop state algorithm of the specific user equipment in real time, the effectiveness of the algorithm is improved, and the corresponding labor cost can be effectively reduced.
Optionally, extracting features of waveform data according to a fixed time period to form a feature time sequence, where the waveform data is data collected by a bus side at high frequency, and the method includes: setting a characteristic calculation time window of the waveform data, and carrying out split-phase characteristic calculation; for high-count users, calculating AC two-phase load characteristics to form a characteristic time sequence; for high-supply low-count users, calculating ABC three-phase load characteristics to form a characteristic time sequence; the characteristic time sequence comprises a three-phase current effective value, a three-phase current 3 rd harmonic content, a three-phase current 5 th harmonic content, a three-phase current 7 th harmonic content, a three-phase current 9 th harmonic content, a three-phase current 11 th harmonic content, a three-phase power factor and a three-phase voltage distortion rate.
Optionally, the data preprocessing is performed on the characteristic time sequence, including: determining a maximum threshold value in each characteristic dimension of a preset characteristic time sequence; when the characteristic of the characteristic time sequence exceeds the maximum threshold value, replacing the characteristic of the characteristic time sequence by the characteristic in the previous unit time corresponding to the characteristic time sequence; and carrying out median filtering on the characteristic time sequence according to characteristic dimensions, reserving mutation points, and finishing data preprocessing.
Optionally, based on DBSCAN clustering, clustering the preprocessed feature timing sequence to obtain a clustering result, extracting a sequence point of the preprocessed feature timing sequence according to the clustering result, reconstructing a feature difference timing sequence, and determining an optimized feature difference timing sequence, including: based on DBSCAN clustering, clustering the preprocessed characteristic time sequence to obtain a clustering result, wherein the clustering result comprises minimum neighborhood points and neighborhood radii; calibrating the preprocessed characteristic time sequence points according to the clustering result, and only reserving the starting points and the ending points belonging to the same class in the preprocessed characteristic time sequence; sequentially calculating the characteristic difference values among different categories according to time sequence to obtain a characteristic difference value time sequence, and calculating the time difference value of category mutation as the characteristic quantity of the newly added transient time in the characteristic time sequence; and determining a three-phase current effective value/power minimum judgment threshold value of preset equipment, sequentially combining partial differences in the characteristic difference time sequence according to the three-phase current effective value/power minimum judgment threshold value, and determining an optimized characteristic difference time sequence.
Optionally, performing difference matching on the optimized feature difference time sequence based on an optimal search algorithm, and determining a difference matching result includes: based on an optimal searching method, performing difference matching on the optimal characteristic difference time sequence according to a priority matching single difference principle, a priority matching earliest moment principle and a continuous difference redundancy searching principle to finish matched difference combination; if the difference combination is a non-single positive and negative difference combination, performing class difference matching, and determining a difference matching result; if the difference combination is a single positive and negative difference combination, no class difference matching is performed.
Optionally, mapping a start-stop state event belonging to the same device based on a difference matching result, optimizing a feature dimension, identifying an electrical type of the device, determining a load identification result, and establishing a proprietary model feature library of industrial and commercial users, including: classifying the difference matching result according to preset identity judgment thresholds of all characteristic dimensions, and determining the variation mean value and the deviation coefficient of all characteristic quantities in the start-stop transient process of the same equipment; according to the time mark corresponding to the starting transient process, extracting a characteristic time sequence before filtering, calculating an impact parameter in the starting transient process as a newly added characteristic quantity, and adopting the calculation formula as follows:
Wherein, M MAX and T MAX are the maximum value and the corresponding maximum time point of various characteristics in the starting transient process, and M Steady and T Steady are various characteristic values and time points corresponding to the end of the transient process;
based on univariate feature selection and a Pearson correlation coefficient method, weight ordering of feature quantities is obtained, feature selection is carried out according to sequence, features capable of identifying the most devices are selected until the selected feature combination can complete identification of all devices, identifiers are marked on the selected features, and a load identification result is determined; and according to the actual use function requirement of the user, extracting comprehensive characteristics based on basic characteristics, marking identifiers on corresponding characteristics for function application, combining the characteristics for identification, and deleting useless characteristics.
Optionally, for bus-side waveform data of a period to be analyzed, according to the load identification result, obtaining the duty ratio of load power consumption of various devices of the period, to realize the decomposition of the bus load of industrial and commercial users, including: for bus side waveform data of a period of time to be analyzed, dividing time nodes corresponding to start-stop processes of various devices, and determining a time window, wherein the time nodes only comprise time nodes corresponding to transient start in a transient starting process and time nodes corresponding to transient end in a transient closing process; determining the electricity consumption of various devices corresponding to each time window of the period to be analyzed; adding the electricity consumption of various devices corresponding to each time window of the period to be analyzed to obtain the electricity consumption of various devices in the period to be analyzed; and determining the electricity consumption of various devices in the period of time to be analyzed and the total electricity consumption in the period of time to be analyzed, and obtaining the duty ratio of the load power consumption of various devices.
Optionally, determining the electricity consumption of each type of device corresponding to each time window of the period to be analyzed includes: when the time window corresponds to a single device, determining that the power consumption of the single device is discrete integration of the three-phase active power value on the time window, and determining a calculation formula of the power consumption of the single device as follows:
wherein W is the power consumption of the single device, Δt is a fixed time period, and p Ai、pBi、pCi is the corresponding three-phase active power value in the time window;
and determining the average active power value of the single equipment, wherein the calculation formula is as follows:
Wherein, For the average active power value of the single device, P i is the active power value of the single device, T M is the corresponding time point at the end of the transient, and T N is the steady-state time point farthest from the transient.
Optionally, determining the electricity consumption of each type of device corresponding to each time window of the period to be analyzed includes: when the time window corresponds to a combined device, determining that the electricity consumption of the combined device is discrete integration of three-phase average active power values after the combined device singly runs into a steady state on the time window, and calculating an average value in other time windows corresponding to the state change of the device by the average active power values, wherein the average value is calculated by a formula:
Wherein P j is the active power value calculated in each time window where a state change occurs, And (3) calculating the average active power value of j-class equipment in each time window with state change, wherein T i is the corresponding steady-state calculation time.
In accordance with another aspect of the present embodiment, a system 600 for load splitting for business users is also provided. Referring to fig. 6, the system 600 further includes: the characteristic time sequence forming module 610 is configured to extract characteristics of waveform data according to a fixed time period, and form a characteristic time sequence, where the waveform data is data collected by a bus side at high frequency; a module 620 for determining a preprocessed feature timing sequence, configured to perform data preprocessing on the feature timing sequence, and determine a preprocessed feature timing sequence; the module 630 for determining an optimized feature difference time sequence is configured to perform clustering operation on the preprocessed feature time sequence based on DBSCAN clustering to obtain a clustering result, extract sequence points of the preprocessed feature time sequence according to the clustering result, reconstruct a feature difference time sequence, and determine an optimized feature difference time sequence; the difference value matching result determining module 640 is configured to perform difference value matching on the optimized feature difference value time sequence based on an optimal search algorithm, and determine a difference value matching result; the load identification result determining module 650 is configured to map start-stop state events belonging to the same device based on the difference matching result, optimize feature dimensions, identify an electrical type of the device, determine a load identification result, and establish a proprietary model feature library of industrial and commercial users; and the obtained equipment load power consumption duty ratio module 660 is used for obtaining the duty ratio of various equipment loads in a period of time according to the load identification result for the bus side waveform data in the period of time to be analyzed, and realizing the decomposition of the industrial and commercial user bus loads.
Optionally, forming the feature timing sequence module 610 includes: the phase separation characteristic calculation sub-module is used for setting a characteristic calculation time window of the waveform data and carrying out phase separation characteristic calculation; the sub-module for forming two-phase characteristic time sequence is used for calculating AC two-phase load characteristics for high-count users to form a characteristic time sequence; the sub-module for forming three-phase characteristic time sequence is used for calculating ABC three-phase load characteristics for high-supply low-count users to form a characteristic time sequence; the characteristic time sequence comprises a three-phase current effective value, a three-phase current 3 rd harmonic content, a three-phase current 5 th harmonic content, a three-phase current 7 th harmonic content, a three-phase current 9 th harmonic content, a three-phase current 11 th harmonic content, a three-phase power factor and a three-phase voltage distortion rate.
Optionally, the determining preprocessed feature timing sequence module 620 includes: the maximum threshold determining sub-module is used for determining a maximum threshold value in each characteristic dimension of a preset characteristic time sequence; a substitute feature sub-module, configured to replace, when a feature of the feature timing sequence exceeds the maximum threshold, a feature of the feature timing sequence with a feature in a previous unit time corresponding to the feature timing sequence; and the preprocessing sub-module is used for carrying out median filtering on the characteristic time sequence according to the characteristic dimension, reserving mutation points and finishing data preprocessing.
Optionally, the determine optimized feature difference timing sequence module 630 includes: the cluster result obtaining sub-module is used for carrying out cluster operation on the preprocessed characteristic time sequence based on DBSCAN cluster to obtain a cluster result, wherein the cluster result comprises minimum neighborhood point number and neighborhood radius; the feature time Xu Xulie point sub-module is used for calibrating the feature time sequence points after pretreatment according to the clustering result, and only preserving the starting points and the ending points belonging to the same class in the feature time sequence after pretreatment; the sub-module of the sequence of the obtained characteristic difference is used for calculating the characteristic difference among different categories sequentially according to the time sequence, obtaining the sequence of the characteristic difference, and calculating the time difference with abrupt change of category as the characteristic quantity of the newly added transient time in the sequence of the characteristic difference; the sub-module for determining the optimized characteristic difference time sequence is used for determining a three-phase current effective value/power minimum judging threshold value of preset equipment, and sequentially combining partial differences in the characteristic difference time sequence according to the three-phase current effective value/power minimum judging threshold value to determine the optimized characteristic difference time sequence.
Optionally, the determining difference matching result module 640 includes: the matching difference value combination sub-module is used for carrying out difference value matching on the optimized characteristic difference value time sequence according to a principle of preferentially matching a single difference value source with the earliest moment and a principle of continuously searching difference value redundancy based on an optimal searching method, so as to complete matched difference value combination; the difference value matching result determining submodule is used for performing class difference value matching if the difference value combination is a non-single positive and negative difference value combination, and determining a difference value matching result; and the class-difference matching sub-module is not used for not carrying out class-difference matching if the difference combination is a single positive-negative difference combination.
Optionally, determining the load recognition result module 650 includes: the sub-module for determining the change mean value and the deviation coefficient of each characteristic quantity is used for classifying the difference matching result according to the preset identity judgment threshold value of each characteristic dimension and determining the change mean value and the deviation coefficient of each characteristic quantity in the start-stop transient process of the same equipment; the impact parameter calculating sub-module is used for extracting a characteristic time sequence before filtering according to a time mark corresponding to the starting transient process, calculating the impact parameter in the starting transient process as a newly added characteristic quantity, and the calculation formula is as follows:
Wherein, M MAX and T MAX are the maximum value and the corresponding maximum time point of various characteristics in the starting transient process, and M Steady and T Steady are various characteristic values and time points corresponding to the end of the transient process;
The load identification result determining sub-module is used for obtaining weight ordering of all feature quantities based on univariate feature selection and Pelson correlation coefficient methods, sequentially selecting the features, selecting the features capable of identifying the most devices until the selected feature combination can complete identification of all devices, marking identifiers on the selected features, and determining a load identification result; and the useless feature deleting sub-module is used for extracting comprehensive features based on basic features according to the actual use function requirements of users, marking identifiers on corresponding features for function application and deleting useless features in combination with the features for identification.
Optionally, the device load power consumption duty cycle module 660 is obtained, including: the time window determining submodule is used for dividing time nodes corresponding to the start-stop processes of various devices for bus side waveform data of a period to be analyzed, and determining a time window, wherein the time nodes only comprise time nodes corresponding to the transient start in the transient starting process and time nodes corresponding to the transient end in the transient closing process; the power consumption sub-module is used for determining the time windows of various devices and is used for determining the power consumption of various devices corresponding to the time windows of the period to be analyzed; the power consumption sub-module is used for adding the power consumption of each device corresponding to each time window of the period to be analyzed to obtain the power consumption of each device in the period to be analyzed; and obtaining a duty ratio submodule of various equipment loads, wherein the duty ratio submodule is used for determining the power consumption of various equipment in a period to be analyzed and the total power consumption in the period to be analyzed to obtain the duty ratio of various equipment loads.
Optionally, the power consumption submodule for determining the time windows of various devices includes: and the power consumption unit is used for determining that the power consumption of the single device is the discrete integral of the three-phase active power value on the time window when the time window corresponds to the single device, and the calculation formula for determining the power consumption of the single device is as follows:
wherein W is the power consumption of the single device, Δt is a fixed time period, and p Ai、pBi、pCi is the corresponding three-phase active power value in the time window;
the unit for determining the average active power value of the single equipment is used for determining the average active power value of the single equipment, and the calculation formula is as follows:
Wherein, For the average active power value of the single device, P i is the active power value of the single device, T M is the corresponding time point at the end of the transient, and T N is the steady-state time point farthest from the transient.
Optionally, the power consumption submodule for determining the time windows of various devices includes: when the time window corresponds to a combined device, determining the electricity consumption of the combined device, wherein the electricity consumption of the combined device is the discrete integration of the three-phase average active power value after the combined device singly runs into a steady state on the time window, the average active power value is obtained by calculating the average value in other time windows corresponding to the state change of the device, and the calculation formula is as follows:
Wherein P j is the active power value calculated in each time window where a state change occurs, And (3) calculating the average active power value of j-class equipment in each time window with state change, wherein T i is the corresponding steady-state calculation time.
A system 600 for load splitting for business users according to an embodiment of the present invention corresponds to a method 100 for load splitting for business users according to another embodiment of the present invention, and is not described herein.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The scheme in the embodiment of the application can be realized by adopting various computer languages, such as object-oriented programming language Java, an transliteration script language JavaScript and the like.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (12)
1. A method for load splitting for business users, comprising:
Extracting characteristics of waveform data according to a fixed time period to form a characteristic time sequence, wherein the waveform data is data acquired by a bus side at high frequency;
performing data preprocessing on the characteristic time sequence, and determining the preprocessed characteristic time sequence;
Based on DBSCAN clustering, clustering the preprocessed feature time sequence to obtain a clustering result, extracting sequence points of the preprocessed feature time sequence according to the clustering result, reconstructing a feature difference time sequence, and determining an optimized feature difference time sequence;
Performing difference matching on the optimized characteristic difference time sequence based on an optimal search algorithm, and determining a difference matching result;
mapping start-stop state events belonging to the same equipment based on a difference matching result, optimizing feature dimensions, identifying the electrical type of the equipment, determining a load identification result, and establishing an exclusive model feature library of industrial and commercial users;
For bus side waveform data to be analyzed for a period of time, according to the load identification result, obtaining the duty ratio of load power consumption of various devices for the period of time, and realizing the decomposition of the bus load of industrial and commercial users;
And for bus side waveform data to be analyzed for a period of time, according to the load identification result, obtaining the duty ratio of load power consumption of various devices for the period of time, and realizing the decomposition of the bus load of industrial and commercial users, wherein the method comprises the following steps: for bus side waveform data of a period of time to be analyzed, dividing time nodes corresponding to start-stop processes of various devices, and determining a time window, wherein the time nodes only comprise time nodes corresponding to transient start in a transient starting process and time nodes corresponding to transient end in a transient closing process; determining the electricity consumption of various devices corresponding to each time window of the period to be analyzed; adding the electricity consumption of various devices corresponding to each time window of the period to be analyzed to obtain the electricity consumption of various devices in the period to be analyzed; determining the electricity consumption of various devices within a period of time to be analyzed and the total electricity consumption within the period of time to be analyzed to obtain the duty ratio of the load power consumption of various devices;
Determining the electricity consumption of various devices corresponding to each time window of the period to be analyzed, including: when the time window corresponds to a single device, determining that the power consumption of the single device is discrete integration of the three-phase active power value on the time window, and determining a calculation formula of the power consumption of the single device as follows:
wherein W is the electricity consumption of the single device, deltat is a fixed time period, and p Ai、pBi、pCi is the corresponding three-phase active power value in a time window;
and determining the average active power value of the single equipment, wherein the calculation formula is as follows:
Wherein, For the average active power value of the single device, P i is the active power value of the single device, T M is the corresponding time point at the end of the transient process, and T N is the steady-state time point farthest from the transient process;
Determining the electricity consumption of various devices corresponding to each time window of the period to be analyzed, including: when the time window corresponds to a combined device, determining that the electricity consumption of the combined device is discrete integration of three-phase average active power values after the combined device singly runs into a steady state on the time window, and calculating an average value in other time windows corresponding to the state change of the device by the average active power values, wherein the average value is calculated by a formula:
Wherein P j is the active power value calculated in each time window where a state change occurs, And (3) calculating the average active power value of j-class equipment in each time window with state change, wherein T i is the corresponding steady-state calculation time.
2. The method of claim 1, wherein the extracting features of waveform data according to a fixed time period to form a feature timing sequence, the waveform data being data collected by a bus side at a high frequency, comprises:
setting a characteristic calculation time window of the waveform data, and carrying out split-phase characteristic calculation;
for high-count users, calculating AC two-phase load characteristics to form a characteristic time sequence;
for high-supply low-count users, calculating ABC three-phase load characteristics to form a characteristic time sequence;
the characteristic time sequence comprises a three-phase current effective value, a three-phase current 3 rd harmonic content, a three-phase current 5 th harmonic content, a three-phase current 7 th harmonic content, a three-phase current 9 th harmonic content, a three-phase current 11 th harmonic content, a three-phase power factor and a three-phase voltage distortion rate.
3. The method of claim 1, wherein the data preprocessing of the signature timing sequence comprises:
Determining a maximum threshold value in each characteristic dimension of a preset characteristic time sequence;
When the characteristic of the characteristic time sequence exceeds the maximum threshold value, replacing the characteristic of the characteristic time sequence by the characteristic in the previous unit time corresponding to the characteristic time sequence;
and carrying out median filtering on the characteristic time sequence according to characteristic dimensions, reserving mutation points, and finishing data preprocessing.
4. The method of claim 1, wherein clustering the preprocessed feature timing sequence based on DBSCAN clustering to obtain a clustering result, extracting sequence points of the preprocessed feature timing sequence based on the clustering result, reconstructing a feature difference timing sequence, and determining an optimized feature difference timing sequence, comprises:
Based on DBSCAN clustering, clustering the preprocessed characteristic time sequence to obtain a clustering result, wherein the clustering result comprises minimum neighborhood points and neighborhood radii;
calibrating the preprocessed characteristic time sequence points according to the clustering result, and only reserving the starting points and the ending points belonging to the same class in the preprocessed characteristic time sequence;
Sequentially calculating the characteristic difference values among different categories according to time sequence to obtain a characteristic difference value time sequence, and calculating the time difference value of category mutation as the characteristic quantity of the newly added transient time in the characteristic time sequence;
and determining a three-phase current effective value/power minimum judgment threshold value of preset equipment, sequentially combining partial differences in the characteristic difference time sequence according to the three-phase current effective value/power minimum judgment threshold value, and determining an optimized characteristic difference time sequence.
5. The method of claim 1, wherein performing a difference match on the optimized feature difference timing sequence based on an optimal search algorithm, determining a difference match result, comprises:
Based on an optimal searching method, performing difference matching on the optimal characteristic difference time sequence according to a priority matching single difference principle, a priority matching earliest moment principle and a continuous difference redundancy searching principle to finish matched difference combination;
If the difference combination is a non-single positive and negative difference combination, performing class difference matching, and determining a difference matching result;
If the difference combination is a single positive and negative difference combination, no class difference matching is performed.
6. The method of claim 1, wherein mapping start-stop status events attributed to the same device based on the difference matching results, optimizing feature dimensions, identifying electrical types of the device, determining load identification results, and establishing a proprietary model feature library for business users comprises:
Classifying the difference matching result according to preset identity judgment thresholds of all characteristic dimensions, and determining the variation mean value and the deviation coefficient of all characteristic quantities in the start-stop transient process of the same equipment;
according to the time mark corresponding to the starting transient process, extracting a characteristic time sequence before filtering, calculating an impact parameter in the starting transient process as a newly added characteristic quantity, and adopting the calculation formula as follows:
Wherein, M MAX and T MAX are the maximum value and the corresponding maximum time point of various characteristics in the starting transient process, and M Steady and T Steady are various characteristic values and time points corresponding to the end of the transient process;
based on univariate feature selection and a Pearson correlation coefficient method, weight ranking of each feature quantity is obtained, feature selection is carried out according to sequence, features capable of identifying the most devices are selected until the selected feature combination completes identification of all devices, identifiers are marked on the selected features, and a load identification result is determined, wherein each feature quantity is a feature quantity of newly added transient time in a feature time sequence and a newly added feature quantity;
And according to the actual use function requirement of the user, extracting comprehensive characteristics based on basic characteristics, marking identifiers on corresponding characteristics for function application, combining the characteristics for identification, and deleting useless characteristics.
7. A system for load splitting for business users, comprising:
the characteristic time sequence forming module is used for extracting characteristics of waveform data according to a fixed time period to form a characteristic time sequence, wherein the waveform data is data acquired by a bus side at high frequency;
The characteristic time sequence determining module is used for carrying out data preprocessing on the characteristic time sequence and determining the characteristic time sequence after preprocessing;
the module for determining the optimized characteristic difference time sequence is used for carrying out clustering operation on the preprocessed characteristic time sequence based on DBSCAN clustering to obtain a clustering result, extracting sequence points of the preprocessed characteristic time sequence according to the clustering result, reconstructing the characteristic difference time sequence, and determining the optimized characteristic difference time sequence;
the difference value matching result determining module is used for performing difference value matching on the optimized characteristic difference value time sequence based on an optimal search algorithm, and determining a difference value matching result;
The load identification result determining module is used for mapping start-stop state events belonging to the same equipment based on the difference value matching result, optimizing feature dimensions, identifying the electrical type of the equipment, determining a load identification result and establishing an exclusive model feature library of industrial and commercial users;
The device load power consumption ratio obtaining module is used for obtaining the power consumption ratio of various devices in a period of time according to the load identification result for bus side waveform data in a period of time to be analyzed, and realizing the decomposition of the industrial and commercial user bus load;
Obtaining a device load power consumption duty ratio module, comprising: the time window determining submodule is used for dividing time nodes corresponding to the start-stop processes of various devices for bus side waveform data of a period to be analyzed, and determining a time window, wherein the time nodes only comprise time nodes corresponding to the transient start in the transient starting process and time nodes corresponding to the transient end in the transient closing process; the power consumption sub-module is used for determining the time windows of various devices and is used for determining the power consumption of various devices corresponding to the time windows of the period to be analyzed; the power consumption sub-module is used for adding the power consumption of each device corresponding to each time window of the period to be analyzed to obtain the power consumption of each device in the period to be analyzed; the load power consumption proportion submodule of each type of equipment is obtained and is used for determining the electricity consumption of each type of equipment in a period of time to be analyzed and the total electricity consumption in the period of time to be analyzed to obtain the load power consumption proportion of each type of equipment;
The power consumption submodule for determining the time window of each type of equipment comprises the following steps: and the power consumption unit is used for determining that the power consumption of the single device is the discrete integral of the three-phase active power value on the time window when the time window corresponds to the single device, and the calculation formula for determining the power consumption of the single device is as follows:
wherein W is the electricity consumption of the single device, deltat is a fixed time period, and p Ai、pBi、pCi is the corresponding three-phase active power value in a time window;
the unit for determining the average active power value of the single equipment is used for determining the average active power value of the single equipment, and the calculation formula is as follows:
Wherein, For the average active power value of the single device, P i is the active power value of the single device, T M is the corresponding time point at the end of the transient process, and T N is the steady-state time point farthest from the transient process;
The power consumption submodule for determining the time window of each type of equipment comprises the following steps: when the time window corresponds to a combined device, determining the electricity consumption of the combined device, wherein the electricity consumption of the combined device is the discrete integration of the three-phase average active power value after the combined device singly runs into a steady state on the time window, the average active power value is obtained by calculating the average value in other time windows corresponding to the state change of the device, and the calculation formula is as follows:
Wherein P j is the active power value calculated in each time window where a state change occurs, And (3) calculating the average active power value of j-class equipment in each time window with state change, wherein T i is the corresponding steady-state calculation time.
8. The system of claim 7, wherein forming the feature timing sequence module comprises:
the phase separation characteristic calculation sub-module is used for setting a characteristic calculation time window of the waveform data and carrying out phase separation characteristic calculation;
The sub-module for forming two-phase characteristic time sequence is used for calculating AC two-phase load characteristics for high-count users to form a characteristic time sequence;
The sub-module for forming three-phase characteristic time sequence is used for calculating ABC three-phase load characteristics for high-supply low-count users to form a characteristic time sequence;
the characteristic time sequence comprises a three-phase current effective value, a three-phase current 3 rd harmonic content, a three-phase current 5 th harmonic content, a three-phase current 7 th harmonic content, a three-phase current 9 th harmonic content, a three-phase current 11 th harmonic content, a three-phase power factor and a three-phase voltage distortion rate.
9. The system of claim 7, wherein determining the preprocessed signature timing sequence module comprises:
the maximum threshold determining sub-module is used for determining a maximum threshold value in each characteristic dimension of a preset characteristic time sequence;
A substitute feature sub-module, configured to replace, when a feature of the feature timing sequence exceeds the maximum threshold, a feature of the feature timing sequence with a feature in a previous unit time corresponding to the feature timing sequence;
And the preprocessing sub-module is used for carrying out median filtering on the characteristic time sequence according to the characteristic dimension, reserving mutation points and finishing data preprocessing.
10. The system of claim 7, wherein determining an optimized feature difference timing sequence module comprises:
the cluster result obtaining sub-module is used for carrying out cluster operation on the preprocessed characteristic time sequence based on DBSCAN cluster to obtain a cluster result, wherein the cluster result comprises minimum neighborhood point number and neighborhood radius;
the feature time Xu Xulie point sub-module is used for calibrating the feature time sequence points after pretreatment according to the clustering result, and only preserving the starting points and the ending points belonging to the same class in the feature time sequence after pretreatment;
The sub-module of the sequence of the obtained characteristic difference is used for calculating the characteristic difference among different categories sequentially according to the time sequence, obtaining the sequence of the characteristic difference, and calculating the time difference with abrupt change of category as the characteristic quantity of the newly added transient time in the sequence of the characteristic difference;
The sub-module for determining the optimized characteristic difference time sequence is used for determining a three-phase current effective value/power minimum judging threshold value of preset equipment, and sequentially combining partial differences in the characteristic difference time sequence according to the three-phase current effective value/power minimum judging threshold value to determine the optimized characteristic difference time sequence.
11. The system of claim 7, wherein determining a difference match result module comprises:
the matching difference value combination sub-module is used for carrying out difference value matching on the optimized characteristic difference value time sequence according to a priority matching single difference value principle, a priority matching earliest moment principle and a continuous difference value redundancy searching principle based on an optimal searching method, so as to complete matched difference value combination;
The difference value matching result determining submodule is used for performing class difference value matching if the difference value combination is a non-single positive and negative difference value combination, and determining a difference value matching result;
and the class-difference matching sub-module is not used for not carrying out class-difference matching if the difference combination is a single positive-negative difference combination.
12. The system of claim 7, wherein determining a load recognition result module comprises:
the sub-module for determining the change mean value and the deviation coefficient of each characteristic quantity is used for classifying the difference matching result according to the preset identity judgment threshold value of each characteristic dimension and determining the change mean value and the deviation coefficient of each characteristic quantity in the start-stop transient process of the same equipment;
The impact parameter calculating sub-module is used for extracting a characteristic time sequence before filtering according to a time mark corresponding to the starting transient process, calculating the impact parameter in the starting transient process as a newly added characteristic quantity, and the calculation formula is as follows:
Wherein, M MAX and T MAX are the maximum value and the corresponding maximum time point of various characteristics in the starting transient process, and M Steady and T Steady are various characteristic values and time points corresponding to the end of the transient process;
The load identification result determining sub-module is used for obtaining weight ordering of all feature quantities based on univariate feature selection and Pelson correlation coefficient methods, sequentially selecting the features, selecting the features capable of identifying the most devices until the selected feature combination completes identification of all devices, marking identifiers on the selected features, and determining a load identification result, wherein each feature quantity is a feature quantity of newly added transient time in a feature time sequence and added with a newly added feature quantity;
And the useless feature deleting sub-module is used for extracting comprehensive features based on basic features according to the actual use function requirements of users, marking identifiers on corresponding features for function application and deleting useless features in combination with the features for identification.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110213501.8A CN113326296B (en) | 2021-02-25 | 2021-02-25 | Load decomposition method and system suitable for industrial and commercial users |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110213501.8A CN113326296B (en) | 2021-02-25 | 2021-02-25 | Load decomposition method and system suitable for industrial and commercial users |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113326296A CN113326296A (en) | 2021-08-31 |
| CN113326296B true CN113326296B (en) | 2024-06-07 |
Family
ID=77414423
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110213501.8A Active CN113326296B (en) | 2021-02-25 | 2021-02-25 | Load decomposition method and system suitable for industrial and commercial users |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113326296B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015059272A1 (en) * | 2013-10-24 | 2015-04-30 | Universite Libre De Bruxelles | Improved non-intrusive appliance load monitoring method and device |
| WO2017035964A1 (en) * | 2015-08-31 | 2017-03-09 | 中车大连电力牵引研发中心有限公司 | Method and system for determining load characteristics of electric power system |
| CN106786534A (en) * | 2016-12-28 | 2017-05-31 | 天津求实智源科技有限公司 | A kind of non-intrusive electrical load transient process discrimination method and system |
| CN109387712A (en) * | 2018-10-09 | 2019-02-26 | 厦门理工学院 | Non-intrusion type cutting load testing and decomposition method based on state matrix decision tree |
| CN109782086A (en) * | 2018-12-25 | 2019-05-21 | 武汉中原电子信息有限公司 | A kind of non-intruding load recognition methods based on the analysis of various dimensions signal |
| CN110084719A (en) * | 2019-06-11 | 2019-08-02 | 国网安徽省电力有限公司培训中心 | A kind of distribution network load type device for identifying |
| CN110889465A (en) * | 2019-12-12 | 2020-03-17 | 武汉大学 | Power demand side equipment identification method and system based on self-adaptive resonant network |
| CN110954744A (en) * | 2019-11-18 | 2020-04-03 | 浙江工业大学 | Non-invasive load monitoring method based on event detection |
| CN110956220A (en) * | 2019-12-11 | 2020-04-03 | 深圳市活力天汇科技股份有限公司 | Non-invasive household appliance load identification method |
| CN111244954A (en) * | 2020-03-23 | 2020-06-05 | 广东电科院能源技术有限责任公司 | Non-invasive load identification method and device |
| CN111830347A (en) * | 2020-07-17 | 2020-10-27 | 四川大学 | Two-stage non-invasive load monitoring method based on event |
| CN112198385A (en) * | 2020-09-30 | 2021-01-08 | 国网山西省电力公司晋中供电公司 | Non-invasive load monitoring method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103001230B (en) * | 2012-11-16 | 2014-10-15 | 天津大学 | Non-invasive power load monitoring and decomposing current mode matching method |
| US20140207398A1 (en) * | 2013-01-23 | 2014-07-24 | Samsung Electronics Co., Ltd | Transient Normalization for Appliance Classification, Disaggregation, and Power Estimation in Non-Intrusive Load Monitoring |
| CN108009938B (en) * | 2016-11-02 | 2021-12-03 | 中国电力科学研究院 | System load clustering and load period pattern recognition method based on shape |
| US10636007B2 (en) * | 2017-05-15 | 2020-04-28 | Tata Consultancy Services Limited | Method and system for data-based optimization of performance indicators in process and manufacturing industries |
-
2021
- 2021-02-25 CN CN202110213501.8A patent/CN113326296B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015059272A1 (en) * | 2013-10-24 | 2015-04-30 | Universite Libre De Bruxelles | Improved non-intrusive appliance load monitoring method and device |
| WO2017035964A1 (en) * | 2015-08-31 | 2017-03-09 | 中车大连电力牵引研发中心有限公司 | Method and system for determining load characteristics of electric power system |
| CN106786534A (en) * | 2016-12-28 | 2017-05-31 | 天津求实智源科技有限公司 | A kind of non-intrusive electrical load transient process discrimination method and system |
| CN109387712A (en) * | 2018-10-09 | 2019-02-26 | 厦门理工学院 | Non-intrusion type cutting load testing and decomposition method based on state matrix decision tree |
| CN109782086A (en) * | 2018-12-25 | 2019-05-21 | 武汉中原电子信息有限公司 | A kind of non-intruding load recognition methods based on the analysis of various dimensions signal |
| CN110084719A (en) * | 2019-06-11 | 2019-08-02 | 国网安徽省电力有限公司培训中心 | A kind of distribution network load type device for identifying |
| CN110954744A (en) * | 2019-11-18 | 2020-04-03 | 浙江工业大学 | Non-invasive load monitoring method based on event detection |
| CN110956220A (en) * | 2019-12-11 | 2020-04-03 | 深圳市活力天汇科技股份有限公司 | Non-invasive household appliance load identification method |
| CN110889465A (en) * | 2019-12-12 | 2020-03-17 | 武汉大学 | Power demand side equipment identification method and system based on self-adaptive resonant network |
| CN111244954A (en) * | 2020-03-23 | 2020-06-05 | 广东电科院能源技术有限责任公司 | Non-invasive load identification method and device |
| CN111830347A (en) * | 2020-07-17 | 2020-10-27 | 四川大学 | Two-stage non-invasive load monitoring method based on event |
| CN112198385A (en) * | 2020-09-30 | 2021-01-08 | 国网山西省电力公司晋中供电公司 | Non-invasive load monitoring method |
Non-Patent Citations (6)
| Title |
|---|
| Research on Load Clustering Based on Singular Value Decomposition and K-means Clustering Algorithm;Jiao Wang 等;2020 Asia Energy and Electrical Engineering Symposium (AEEES);20200619;第831-835页 * |
| 基于聚类和关联分析的居民用户非侵入式负荷分解;赵文清;张诗满;李刚;;电力自动化设备;20200604(06);第8-19页 * |
| 基于非侵入式用电数据分解的自适应特征库构建与负荷辨识;武昕;焦点;高宇辰;;电力系统自动化;20191225(04);第101-109页 * |
| 居民负荷特征研究及特征库的建立;祁兵;刘利亚;张瑜;翟峰;杨斌;;东北电力技术;20180620(06);第1-8页 * |
| 智能电网环境下基于大数据挖掘的居民负荷设备识别与负荷建模;杨甲甲;赵俊华;文福拴;董朝阳;薛禹胜;;电力建设;20161201(12);第11-23页 * |
| 非介入式负荷辨识感知技术及其典型应用场景;陈骏星溆;徐先勇;孟军;肖剑;陈卓;;湖南电力;20200425(02);第68-73页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113326296A (en) | 2021-08-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111090643B (en) | Mass electricity consumption data mining method based on data analysis system | |
| CN109446730B (en) | Short-term equipment operation data-based generator set load factor missing value recruitment method | |
| CN112785108A (en) | Power grid operation data correlation analysis method and system based on regulation cloud | |
| CN112184491B (en) | Method, device, computer equipment and storage medium for identifying abnormal data in power grid | |
| CN112532643B (en) | Flow anomaly detection method, system, terminal and medium based on deep learning | |
| CN105045927A (en) | Automatic coding method and system for data of labor, materials and machines of construction project | |
| CN110729718A (en) | Industry user work starting monitoring method based on daily load curve | |
| CN110866841A (en) | Power consumer industry dimension power consumption pattern identification analysis method and system based on double clustering method | |
| CN116500385B (en) | Power transmission network monitoring and checking method, device, equipment and medium | |
| CN110378585A (en) | Electric network fault disposes calculating task layout call method, system and storage medium | |
| CN111597241B (en) | Method, device and equipment for data acquisition | |
| CN117559443A (en) | Ordered power utilization control method for large industrial user cluster under peak load | |
| CN113098888A (en) | Abnormal behavior prediction method, device, equipment and storage medium | |
| CN110287237B (en) | Social network structure analysis based community data mining method | |
| CN116562596A (en) | Retired battery processing method and device, retired battery storage vertical warehouse and medium | |
| CN111931318A (en) | Power supply path analysis method and system based on graph calculation | |
| CN113779328A (en) | Power supply monitoring data integration processing method, system, terminal and storage medium | |
| Gao et al. | Online learning of stable integer variables in unit commitment using internal information | |
| Grigoras et al. | Processing of smart meters data for peak load estimation of consumers | |
| CN113326296B (en) | Load decomposition method and system suitable for industrial and commercial users | |
| CN113094448A (en) | Method and device for analyzing vacant state of house and electronic equipment | |
| CN112614012A (en) | User electricity stealing identification method and device | |
| CN112926777B (en) | Day-ahead load prediction method and system considering commercial building energy consumption behaviors | |
| CN111768066B (en) | Park electric heating load coupling relation analysis method and device based on fusion characteristics | |
| CN113657475A (en) | Classification-based power consumption prediction method |
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 |