CN108594035B - Load detection method and system - Google Patents
Load detection method and system Download PDFInfo
- Publication number
- CN108594035B CN108594035B CN201810246406.6A CN201810246406A CN108594035B CN 108594035 B CN108594035 B CN 108594035B CN 201810246406 A CN201810246406 A CN 201810246406A CN 108594035 B CN108594035 B CN 108594035B
- Authority
- CN
- China
- Prior art keywords
- window
- load
- feature information
- candidate load
- candidate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/25—Arrangements for measuring currents or voltages or for indicating presence or sign thereof using digital measurement techniques
- G01R19/2513—Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
本发明公开了一种负荷检测方法，所述方法包括：按照预设时间周期对获取的负荷的总入口处的总线上的电压信号和电流信号进行加窗处理，并确定每个窗口对应的功率信号；根据所述每个窗口对应的功率信号的持续时间特征确定每个窗口的候选负荷集；根据所述每个窗口的功率信号和对应的每个候选负荷的功率信号确定确定每个窗口对应的每个候选负荷的特征信息；根据所述特征信息和每个候选负荷对应的特征权重组确定每个窗口的负荷。本发明采用离线的方式采集数据，使数据更加准确，能够得到更加完整的波形；将开关事件检测和负荷识别放在一起，利用不同的负荷的特征权重和特征信息确定窗口的负荷，提高了负载事件监测的准确度和负荷的识别分类的准确度。
The invention discloses a load detection method. The method includes: performing windowing processing on the voltage signal and the current signal on the bus at the general entrance of the obtained load according to a preset time period, and determining the power corresponding to each window. signal; determine the candidate load set of each window according to the duration characteristic of the power signal corresponding to each window; determine and determine the corresponding power of each window according to the power signal of each window and the corresponding power signal of each candidate load The characteristic information of each candidate load of , and the load of each window is determined according to the characteristic information and the characteristic weight group corresponding to each candidate load. The invention collects data in an off-line way, so that the data is more accurate and a more complete waveform can be obtained; the switching event detection and load identification are put together, and the characteristic weight and characteristic information of different loads are used to determine the load of the window, which improves the load The accuracy of event monitoring and the accuracy of load identification and classification.
Description
Technical Field
The present invention relates to the field of load monitoring technologies, and in particular, to a load detection method and system.
Background
In order to ensure that the energy consumption requirement of the electric power can be fully met and the residential building and the commercial building can normally operate, effective measures should be taken to reasonably save the electric power and improve the utilization efficiency of the electric power, so that the monitoring and the measurement of the load become more important.
In order to reduce investment and time consumption while monitoring and measuring load, we obtain current and voltage information from buses of residential and commercial buildings, analyze and process the information, and obtain signals of voltage and current. Then, corresponding features are extracted from the signals, and different loads are decomposed by using corresponding algorithms to obtain power utilization information of each load. Since each load has a plurality of functions and a complex principle, it is sometimes difficult to effectively classify the load with continuously changing multiple states by using only voltage and current information.
The current load monitoring method mainly utilizes current and voltage information on a bus to analyze and process, and simply monitors in real time from the aspect of electric signals. The types of the methods are mainly divided into two types, namely an analysis method based on load steady-state characteristics and an analysis method based on load transient characteristics. The analysis method based on the load steady-state characteristics mainly samples and superposes typical current and voltage of a load to realize the simulation of the load steady-state waveform. The load transient-based analysis method mainly samples transient current and voltage, extracts characteristic values such as a waveform coefficient, a peak-to-peak value and the like, and then classifies the characteristic values. The classification method mainly comprises a C4.5 decision tree algorithm, a spectral clustering algorithm and the like, wherein the C4.5 decision tree algorithm is composed of decision nodes, branches and leaf nodes, the decision nodes represent samples, the branches represent different values of a certain decision node, and the leaf nodes represent possible classification results. The process is to train the model by using a training set to obtain a model, then to make a decision from a root node, and to reach a leaf node along a branch to obtain a possible classification label. The spectral clustering algorithm firstly prepares data, generates an adjacency matrix of the graph, then normalizes a Prasiian matrix, generates eigenvalues and eigenvectors, performs clustering, and finally performs classification according to a partition criterion. Although these methods can achieve load recognition to some extent, there are some drawbacks, mainly manifested by the following points: in a general monitoring, detecting and identifying method, a switching event and a load identification are distinguished separately, and the conditions of false detection or non-detection of the switching event and signal interference exist, so that the subsequent identification is influenced; the detection and identification are carried out only from the change of current and voltage, and a plurality of loads are easy to be confused; and the load is monitored in real time, the obtained information quantity is limited, and certain obstacles are caused to the detection of the switching event and the load identification.
Disclosure of Invention
The invention provides a load detection method and a load detection system, which aim to solve the problem of how to detect a load.
In order to solve the above problem, according to an aspect of the present invention, there is provided a load detection method including:
windowing the acquired voltage signal and current signal on the bus at the total inlet of the load according to a preset time period, and determining a power signal corresponding to each window;
determining a candidate load set of each window according to the duration characteristic of the power signal corresponding to each window, wherein each candidate load set comprises at least one candidate load;
determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the corresponding power signal of each candidate load;
and determining the load of each window according to the characteristic information and the characteristic weight set corresponding to each candidate load.
Preferably, wherein the method further comprises:
and filtering the acquired voltage signal and current signal on the bus at the inlet of the load, and converting the filtered voltage signal and current signal into a voltage digital signal and a current digital signal.
Preferably, the acquired voltage signal and current signal on the bus at the inlet of the load are filtered by a 24-bit analog-to-digital converter with model number AD 1256.
Preferably, wherein the method further comprises:
before determining the candidate load set of each window, determining the candidate load range corresponding to each window according to the power characteristics of the power signal corresponding to each window.
Preferably, wherein the characteristic information includes: edge feature information, trend feature information, temporal feature information, frequency feature information, and sequence feature information.
Preferably, the determining the feature information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each candidate load corresponding to each window includes:
dividing the number of the edge types in each window with the number of the edge types of each corresponding candidate load to determine the edge characteristic information of each candidate load corresponding to each window;
dividing the number of the trend types in each window with the number of the trend types of each corresponding candidate load to determine the trend characteristic information of each candidate load corresponding to each window;
comparing the time interval of each window with the opening time interval of each corresponding candidate load, and determining the time characteristic information of each candidate load corresponding to each window; if the time interval of the window is within the opening time interval of each corresponding candidate load, the time characteristic information is 1, otherwise, the time characteristic information is 0;
dividing the number of the edge type repetition of each window with the number of the edge type repetition of each corresponding candidate load, and determining the frequency characteristic information of each candidate load corresponding to each window;
and subtracting the ratio of the reverse order variation of the candidate load and the sequence variable in the window by using 1 to determine the sequence characteristic information of each candidate load corresponding to each window.
Preferably, the determining the load of each window according to the feature information and the feature weight set corresponding to each candidate load includes:
and respectively forming a characteristic information array corresponding to each candidate load of each window by edge characteristic information, trend characteristic information, time characteristic information, frequency characteristic information and sequence characteristic information, calculating the difference between the multiplied characteristic information array and a characteristic weight set corresponding to each candidate load and a preset critical value, and selecting the candidate load corresponding to the maximum difference as the load of each window.
According to another aspect of the present invention, there is provided a load detection system, the system comprising:
the windowing processing module is used for windowing the acquired voltage signal and current signal on the bus at the total inlet of the load according to a preset time period and determining a power signal corresponding to each window;
a candidate load set determining module, configured to determine a candidate load set of each window according to the duration characteristic of the power signal corresponding to each window, where each candidate load set includes at least one candidate load;
the characteristic information determining module is used for determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each corresponding candidate load;
and the window load determining module is used for determining the load of each window according to the characteristic information and the characteristic weight set corresponding to each candidate load.
Preferably, wherein the system further comprises:
and the filtering processing module is used for filtering the acquired voltage signal and current signal on the bus at the general inlet of the load and converting the voltage signal and the current signal subjected to filtering processing into a voltage digital signal and a current digital signal.
Preferably, the acquired voltage signal and current signal on the bus at the inlet of the load are filtered by a 24-bit analog-to-digital converter with model number AD 1256.
Preferably, wherein the system further comprises:
and the candidate load range determining module is used for determining the candidate load range corresponding to each window according to the power characteristics of the power signal corresponding to each window before determining the candidate load set of each window.
Preferably, wherein the characteristic information includes: edge feature information, trend feature information, temporal feature information, frequency feature information, and sequence feature information.
Preferably, the determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each candidate load corresponding to each window by the characteristic information determining module includes:
an edge feature information determining unit, configured to divide the number of edge types in each window by the number of edge types of each corresponding candidate load, and determine edge feature information of each candidate load corresponding to each window;
the trend characteristic information determining unit is used for dividing the trend type number in each window and the trend type number of each corresponding candidate load to determine the trend characteristic information of each candidate load corresponding to each window;
the time characteristic information determining unit is used for comparing the time interval of each window with the opening time interval of each corresponding candidate load and determining the time characteristic information of each candidate load corresponding to each window; if the time interval of the window is within the opening time interval of each corresponding candidate load, the time characteristic information is 1, otherwise, the time characteristic information is 0;
a frequency characteristic information determining unit, configured to divide the number of edge type repetitions of each window by the number of edge type repetitions of each corresponding candidate load, and determine frequency characteristic information of each candidate load corresponding to each window;
and the sequence characteristic information determining unit is used for determining the sequence characteristic information of each candidate load corresponding to each window by using the ratio of 1 minus the inverse sequence variation of the candidate load and the sequence variable in the window.
Preferably, the window load determining module determines the load of each window according to the feature information and the feature weight set corresponding to each candidate load, and is specifically configured to:
and respectively forming a characteristic information array corresponding to each candidate load of each window by edge characteristic information, trend characteristic information, time characteristic information, frequency characteristic information and sequence characteristic information, calculating the difference between the multiplied characteristic information array and a characteristic weight set corresponding to each candidate load and a preset critical value, and selecting the candidate load corresponding to the maximum difference as the load of each window.
The invention provides a load detection method and a system, which adopt an off-line mode to collect data, so that the data is more accurate, more complete waveforms can be obtained, the identification of loads is not influenced, and more information quantity can be obtained; the power characteristics are adopted to roughly determine the load range, so that the accuracy is improved, and part of workload can be reduced; determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each corresponding candidate load, and putting the switching event detection and the load identification together, thereby reducing the false detection or non-detection condition and improving the identification accuracy; finally, the load of the window is determined by using the characteristic weight and the characteristic information of different loads, so that the method is more representative, and the accuracy of monitoring the load event and the accuracy of identifying and classifying the load are improved.
Drawings
A more complete understanding of exemplary embodiments of the present invention may be had by reference to the following drawings in which:
FIG. 1 is a flow diagram of a load detection method 100 according to an embodiment of the invention;
FIG. 2 is a schematic diagram of an apparatus for signal data processing according to an embodiment of the present invention; and
fig. 3 is a schematic structural diagram of a load detection apparatus 300 according to an embodiment of the present invention.
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 embodiments described herein, which are provided for complete and complete disclosure of the present invention and to fully convey the scope of the present 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, the same units/elements are denoted by the same reference numerals.
Unless otherwise defined, terms (including 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. Further, it will be understood that terms, such as those 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.
Fig. 1 is a flow chart of a load detection method 100 according to an embodiment of the present invention. As shown in fig. 1, in a load detection method 100 according to an embodiment of the present invention, a windowing process is performed on a voltage signal and a current signal on a bus at a total entrance of an acquired load, and a candidate load set of each window is determined according to a duration characteristic of a power signal corresponding to each window; and determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each corresponding candidate load, and determining the load of each window according to the characteristic information. The data are acquired in an off-line mode, so that the data are more accurate, more complete waveforms can be obtained, the load identification is not influenced, and more information quantity can be acquired; the power characteristics are adopted to roughly determine the load range, so that the accuracy is improved, and part of workload can be reduced; determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each corresponding candidate load, and putting the switching event detection and the load identification together, thereby reducing the false detection or non-detection condition and improving the identification accuracy; finally, the load of the window is determined by using the characteristic weight and the characteristic information of different loads, so that the method is more representative, and the accuracy of monitoring the load event and the accuracy of identifying and classifying the load are improved.
The load detection method 100 of the embodiment of the present invention starts at step 101, and performs windowing on the acquired voltage signal and current signal on the bus at the inlet of the load according to a preset time period at step 101, and determines a power signal corresponding to each window.
Preferably, wherein the method further comprises:
and filtering the acquired voltage signal and current signal on the bus at the inlet of the load, and converting the filtered voltage signal and current signal into a voltage digital signal and a current digital signal.
Preferably, the acquired voltage signal and current signal on the bus at the inlet of the load are filtered by a 24-bit analog-to-digital converter with model number AD 1256.
In the embodiment of the invention, the switching event and the load are identified together for judgment, so that the error is reduced. Then, the data and waveform of one day or one week are processed in an off-line mode, so that more information can be obtained, and the result is more accurate. For similar loads, edge features, trend features, time features, sequence features, frequency features and power features are compared, the power features are firstly used for dividing the loads into two types of high power (more than 1000W) and low power (less than 1000W), the identification accuracy can be improved, and the workload during identification can be reduced. And then, effectively identifying by utilizing edge features, trend features, time features, sequence features and frequency features.
Fig. 2 is a schematic diagram of a signal data processing apparatus according to an embodiment of the present invention. As shown in fig. 2, the voltage sensor and the current sensor are respectively used for acquiring a voltage signal and a current signal on a bus at a total inlet of a load, sending the voltage signal and the current signal to the value a/D module, converting the current signal and the voltage signal into a current digital signal and a voltage digital signal, sending the current digital signal and the voltage digital signal to the ARM core module for processing data such as filtering, and then sending the data to the value server through the data transmission module for identification and detection of the load. The device also comprises a storage module used for storing the data in the processing process. A24-bit analog-to-digital converter with the model number of AD1256 is adopted in hardware, and the purpose is mainly to reduce signal interference, so that the obtained signal is more accurate and the waveform is more ideal.
Preferably, a candidate load set for each window is determined according to the duration characteristics of the power signal corresponding to each window in step 102, and each candidate load set comprises at least one candidate load.
In the implementation mode of the invention, according to the collocation of different rising edges and falling edges in each window, the duration of the load is recorded, a classification and identification table is established, the condition which does not meet the duration is deleted, and the rest forms the candidate load set corresponding to each window. For example, if a rising edge and a falling edge are separated by 2min, the duration is too short for the kettle and the refrigerator, so the kettle and the refrigerator corresponding to the candidate window should be deleted.
Preferably, wherein the method further comprises:
before determining the candidate load set of each window, determining the candidate load range corresponding to each window according to the power characteristics of the power signal corresponding to each window. In an embodiment of the present invention, within the detection window of the power waveform, the value Δ α of the rising edge transition is the power characteristic. The power characteristics of the power signal corresponding to each window are extracted and compared with 1000W, the possible load of the window is judged to be a high-power load or a low-power load, the load range is roughly divided, and the workload can be effectively reduced.
Preferably, in step 103, the characteristic information of each candidate load corresponding to each window is determined according to the power signal of each window and the power signal of each candidate load corresponding to each window.
Preferably, wherein the characteristic information includes: edge feature information, trend feature information, temporal feature information, frequency feature information, and sequence feature information.
Preferably, the determining the feature information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each candidate load corresponding to each window includes:
dividing the number of the edge types in each window with the number of the edge types of each corresponding candidate load to determine the edge characteristic information of each candidate load corresponding to each window;
dividing the number of the trend types in each window with the number of the trend types of each corresponding candidate load to determine the trend characteristic information of each candidate load corresponding to each window;
comparing the time interval of each window with the opening time interval of each corresponding candidate load, and determining the time characteristic information of each candidate load corresponding to each window; if the time interval of the window is within the opening time interval of each corresponding candidate load, the time characteristic information is 1, otherwise, the time characteristic information is 0;
dividing the repeated number of the edge types of each window with the number of the trend types of each corresponding candidate load, and determining the frequency characteristic information of each candidate load corresponding to each window;
and subtracting the ratio of the reverse order variation of the candidate load and the sequence variable in the window by using 1 to determine the sequence characteristic information of each candidate load corresponding to each window.
Preferably, the load of each window is determined in step 104 according to the feature information and the feature information weight set corresponding to each candidate load.
Preferably, the determining the load of each window according to the feature information and the feature information weight group corresponding to each candidate load includes:
and respectively forming a characteristic information array corresponding to each candidate load of each window by edge characteristic information, trend characteristic information, time characteristic information, frequency characteristic information and sequence characteristic information, calculating the difference between the characteristic information array and a preset critical value after multiplying the characteristic information array and a characteristic information weight group corresponding to each candidate load, and selecting the candidate load corresponding to the maximum difference as the load of each window.
In an embodiment of the present invention, determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each candidate load corresponding to each window includes: edge feature information, trend feature information, time feature information, frequency feature information and sequence feature information, and then a feature information array X is formed by the feature information. Wherein the power signal for each candidate load is stored in a pre-designed database.
Edge feature information XedgeBy detecting the number N 'of edge types in the window'edgeAnd the number of edge types N of the candidate loadedgeAnd (4) dividing to obtain the final product.
The trend characteristic information, including increasing peak, decreasing peak, pulse, fluctuation, rapid change, gradual decrease and smoothing, is detected through the window. Trend characteristic information XtrendBy number of trend types N 'in detection window'trendAnd the number N of candidate load trend typestrendAnd (4) dividing to obtain the final product.
The time characteristic information refers to comparing the time interval of each window with the opening time interval of each corresponding candidate load to determine the time characteristic information of each candidate load corresponding to each window; if the time interval of the window is within the opening time interval of each corresponding candidate load, the time characteristic information is 1, otherwise, the time characteristic information is 0. According to the candidate window, each load has a time period of opening, for example, the electric cooker is switched on at the time of 6:00-9:00, 11:00-14:00, 16: the three time periods of 00-20:00 are frequently used. If the candidate window is within the time period corresponding to the load, XtimeIs 1, otherwise is 0.
Frequency characteristic information XrateNumber N 'repeated by edge type in detection window'rateNumber of repetitions to candidate load edge type NrateAnd (4) dividing to obtain the final product.
The sequence feature information is determined using a ratio of 1 minus the amount of inverse sequence variation of the candidate payload and the sequence variable in the window. For the electric appliances with fixed sequences, such as washing machines, dish washing machines and the like, a sequence calibration method is adopted. For example, the washing machine has four processes of water injection, soaking, washing and spin-drying, corresponding to 1, 2, 3 and 4, the inverse sequence variation M is calculated as |1-4| + |2-3| + |3-2| + |4-1| 8, and then the sequence variation N in the window is calculated, if the sequence in the window is 2, 1, 3 and 4, the variation N is |1-2| + |2-1| + |3-3| + |4-4| 2. Last XorderThe ratio of M to N is subtracted from 1.
In an embodiment of the present invention, the load of each window is determined according to the feature information and a feature information weight set corresponding to each candidate load. And putting the edge characteristic information, the trend characteristic information, the time characteristic information, the frequency characteristic information and the sequence characteristic information into a characteristic information array X, multiplying the characteristic information array X by the weight omega corresponding to each candidate load to obtain omega TX, wherein the weights of the group corresponding to each candidate load are different for different loads. For example, the sequence characteristic information accounts for a large proportion of fixed loads such as washing machines, the frequency characteristic information accounts for a small proportion of repeated loads such as hot water kettles, and the time characteristic information provides suggestions for most of life loads. And multiplying the feature information array X by the feature information weight group corresponding to each candidate load, calculating a difference value between the feature information array X and a preset critical value lambda, and selecting the candidate load corresponding to the maximum difference value as the load of each window.
Fig. 3 is a schematic structural diagram of a load detection apparatus 300 according to an embodiment of the present invention. As shown in fig. 3, a load detection system 300 according to an embodiment of the present invention includes: a windowing processing module 301, a candidate load set determining module 302, a feature information determining module 303 and a window load determining module 304. Preferably, in the windowing processing module 301, the obtained voltage signal and current signal on the bus at the inlet of the load are windowed according to a preset time period, and a power signal corresponding to each window is determined.
Preferably, wherein the system further comprises:
and the filtering processing module is used for filtering the acquired voltage signal and current signal on the bus at the general inlet of the load and converting the voltage signal and the current signal subjected to filtering processing into a voltage digital signal and a current digital signal.
Preferably, the acquired voltage signal and current signal on the bus at the inlet of the load are filtered by a 24-bit analog-to-digital converter with model number AD 1256.
Preferably, in the candidate load set determining module 302, the candidate load set of each window is determined according to the duration characteristic of the power signal corresponding to each window, and each candidate load set includes at least one candidate load.
Preferably, wherein the system further comprises:
and the candidate load range determining module is used for determining the candidate load range corresponding to each window according to the power characteristics of the power signal corresponding to each window before determining the candidate load set of each window.
Preferably, in the characteristic information determining module 303, the characteristic information of each candidate load corresponding to each window is determined according to the power signal of each window and the power signal of each candidate load corresponding to each window. Preferably, wherein the characteristic information includes: edge feature information, trend feature information, temporal feature information, frequency feature information, and sequence feature information.
Preferably, the determining the characteristic information of each candidate load corresponding to each window according to the power signal of each window and the power signal of each candidate load corresponding to each window by the characteristic information determining module includes:
an edge feature information determining unit, configured to divide the number of edge types in each window by the number of edge types of each corresponding candidate load, and determine edge feature information of each candidate load corresponding to each window;
the trend characteristic information determining unit is used for dividing the trend type number in each window and the trend type number of each corresponding candidate load to determine the trend characteristic information of each candidate load corresponding to each window;
the time characteristic information determining unit is used for comparing the time interval of each window with the opening time interval of each corresponding candidate load and determining the time characteristic information of each candidate load corresponding to each window; if the time interval of the window is within the opening time interval of each corresponding candidate load, the time characteristic information is 1, otherwise, the time characteristic information is 0;
a frequency characteristic information determining unit, configured to divide the number of edge type repetitions of each window by the number of edge types of each corresponding candidate load, and determine frequency characteristic information of each candidate load corresponding to each window;
and the sequence characteristic information determining unit is used for determining the sequence characteristic information of each candidate load corresponding to each window by using the ratio of 1 minus the inverse sequence variation of the candidate load and the sequence variable in the window.
Preferably, in the window load determining module 304, the load of each window is determined according to the feature information and the feature weight set corresponding to each candidate load.
Preferably, the window load determining module determines the load of each window according to the feature information and the feature weight set corresponding to each candidate load, and is specifically configured to:
and respectively forming a characteristic information array corresponding to each candidate load of each window by edge characteristic information, trend characteristic information, time characteristic information, frequency characteristic information and sequence characteristic information, calculating the difference between the multiplied characteristic information array and a characteristic weight set corresponding to each candidate load and a preset critical value, and selecting the candidate load corresponding to the maximum difference as the load of each window.
The load detection apparatus 300 according to the embodiment of the present invention corresponds to the method 100 for detecting a load according to another embodiment of the present invention, and is not described herein again.
The invention has been described with reference to a few embodiments. However, other embodiments of the invention than the one disclosed above are equally possible within the scope of the invention, as would be apparent to a person skilled in the art from the appended patent claims.
Generally, all terms used in the claims are to be interpreted according to their ordinary meaning in the technical field, unless explicitly defined otherwise herein. All references to "a/an/the [ device, component, etc ]" are to be interpreted openly as referring to at least one instance of said device, component, etc., unless explicitly stated otherwise. The steps of any method disclosed herein do not have to be performed in the exact order disclosed, unless explicitly stated.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810246406.6A CN108594035B (en) | 2018-03-23 | 2018-03-23 | Load detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810246406.6A CN108594035B (en) | 2018-03-23 | 2018-03-23 | Load detection method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108594035A CN108594035A (en) | 2018-09-28 |
| CN108594035B true CN108594035B (en) | 2021-03-09 |
Family
ID=63627359
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810246406.6A Active CN108594035B (en) | 2018-03-23 | 2018-03-23 | Load detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108594035B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110187167B (en) * | 2019-04-30 | 2021-06-11 | 广东石油化工学院 | Method and device for detecting load switch event based on manifold classification |
| CN112883545B (en) * | 2021-01-13 | 2022-04-08 | 吉林大学 | A Simulation Method of Power Load Waveform |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106483370B (en) * | 2016-10-21 | 2019-05-03 | 威胜集团有限公司 | Non-intrusion type household loads real-time identification method and device based on multi-feature fusion |
| CN106680637A (en) * | 2017-01-12 | 2017-05-17 | 威胜集团有限公司 | Monitoring method of non-invasive household load switching event in real time |
| CN106908671A (en) * | 2017-01-13 | 2017-06-30 | 中国海洋大学 | A kind of non-intrusion type household loads intelligent detecting method and system |
| CN107561354B (en) * | 2017-08-17 | 2019-09-10 | 江苏智臻能源科技有限公司 | More electric appliance non-intruding discrimination methods based on power duration characteristic |
-
2018
- 2018-03-23 CN CN201810246406.6A patent/CN108594035B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108594035A (en) | 2018-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110956220B (en) | Non-invasive household appliance load identification method | |
| CN110504679B (en) | KM matching algorithm-based non-invasive load identification method | |
| CN116304766B (en) | Multi-sensor-based quick assessment method for state of switch cabinet | |
| Ghosh et al. | An improved load feature extraction technique for smart homes using fuzzy-based NILM | |
| CN106908671A (en) | A kind of non-intrusion type household loads intelligent detecting method and system | |
| CN106569129B (en) | Motor safety detection method and system and electric machine control system | |
| CN102566555B (en) | Major appliance work state monitoring method based on pattern recognition | |
| CN108594035B (en) | Load detection method and system | |
| CN111191908B (en) | Load event detection method, system and storage medium | |
| CN109470957B (en) | Non-invasive electrical appliance identification method based on Frey's distance | |
| CN108470570B (en) | Abnormal sound detection method for motor | |
| CN113567794A (en) | Indoor charging identification method and system for electric bicycles based on dynamic time warping | |
| CN106443164A (en) | Transient state event and steady state event monitoring method during electric energy classification metering process | |
| CN110416995A (en) | A non-invasive load decomposition method and device | |
| CN112952827A (en) | Non-invasive full-load identification technology for accurately identifying charging of electric bicycle | |
| CN111191671A (en) | Electrical appliance waveform detection method and system, electronic equipment and storage medium | |
| CN115128345A (en) | Power grid safety early warning method and system based on harmonic monitoring | |
| CN117807551B (en) | Heart rate abnormality capturing method and system based on intelligent ring | |
| CN111917114B (en) | Power load event detection method suitable for embedded platform | |
| CN115333102B (en) | Multi-time scale adaptive detection method for electric load power pattern | |
| CN118332519A (en) | Distributed photovoltaic anomaly detection method and device based on split-level cyclic neural network | |
| CN115112989B (en) | Non-invasive load monitoring method based on low-frequency data | |
| CN116881773A (en) | Equipment fault recognition method and system based on deep learning | |
| CN112711613B (en) | Electric appliance preference analysis method based on intelligent judgment of power utilization behavior | |
| CN115825602A (en) | A load identification method and system for comprehensive multivariate information similarity analysis |
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 |