Classification and identification method and system for residential electricity utilization facilities, equipment and storage medium

Abstract

The invention discloses a classification and identification method and system for residential electricity utilization facilities, equipment and a storage medium, and the method comprises the steps: firstly extracting current time domain features and current frequency domain features based on a current curve, so as to construct a current feature vector of each recognized residential electricity utilization facility and a current feature vector of a to-be-recognized residential electricity utilization facility; according to the method, different current characteristics of different residential electricity utilization facilities can be comprehensively described, only current data are collected for analysis, the method has the advantage of being not affected by voltage fluctuation, and accuracy and reliability of collected data are guaranteed. Then, the correlation between the current feature vector of the to-be-identified residential electricity utilization facility and the current feature vectors of all identified residential electricity utilization facilities is calculated, and the higher the correlation between the two current feature vectors means that the closer the current features of the two current feature vectors are; therefore, the identified residential electricity utilization facility with the maximum correlation is used as the classification identification result of the residential electricity utilization facility to be identified, and the real-time classification identification requirement can be well met.

Description

Technical Field

[0001] The present invention relates to the technical field of load classification and identification, and in particular, to a classification and identification method and system for residential electrical facilities, electronic equipment, and a computer-readable storage medium. Background Art

[0002] Accurate classification and identification results of residential power facilities can help power companies and energy suppliers optimize power generation plans and power dispatch plans, which is conducive to reducing energy waste and carbon emissions, and is of great significance to the stable operation and sustainable development of the power system. With the continuous improvement of people's living standards and the continuous development of technology, a large number of new residential power facilities are constantly connected to the power system, which brings new challenges to load classification and identification. At present, the existing load classification and identification methods are usually to extract the electrical quantity characteristics of each load, such as current characteristics, voltage characteristics, power characteristics, etc., and then build a training data set to train the classification model (such as support vector machine model, random forest model, etc.), and then use the trained model to classify and identify the newly connected load. For example, patent CN108491893A discloses a household power load classification and identification method based on a deep forest algorithm, which collects data such as real-time voltage, current, active power, and voltage and current harmonic content of household appliances, and then extracts power factor, upper and lower power limits, multiples of starting impact current and normal operating current, etc. as electrical quantity characteristics, and then builds a training data set to train the deep forest model, and finally uses the trained model to classify and identify the load to be identified. However, due to the large number of residential power facilities and the large amount of data, the training time of the model is long, and it is difficult to achieve rapid iteration of the model, thus making it difficult to meet the real-time classification and recognition needs. In addition, the classification and recognition accuracy of the model depends to a large extent on the accuracy of the collected data. However, since voltage data and power data are greatly affected by voltage fluctuations, especially the voltage fluctuations on the residential power side, the voltage fluctuation range on the residential power side is generally +7% to -10% of the nominal voltage, which greatly affects the accuracy of the collected voltage and power data, resulting in poor classification and recognition accuracy of the model. Summary of the invention

[0003] The present invention provides a classification and identification method and system for residential power facilities, electronic equipment, and a computer-readable storage medium, which can improve the accuracy of load classification and identification, and can also well meet real-time classification and identification needs.

[0004] According to one aspect of the present invention, a method for classifying and identifying residential electrical facilities is provided, comprising: The following content:

[0005] Obtaining current curves of all identified residential power facilities, extracting current time domain features and current frequency domain features of each identified residential power facility, so as to construct a current feature vector of each identified residential power facility;

[0006] Obtaining a current curve of the residential power facility to be identified, extracting the current time domain characteristics and current frequency domain characteristics of the residential power facility to be identified, so as to construct a current feature vector of the residential power facility to be identified;

[0007] The correlation between the current feature vector of the residential power facility to be identified and the current feature vectors of all identified residential power facilities is calculated, and the identified residential power facility with the greatest correlation is taken as the classification and identification result of the residential power facility to be identified.

[0008] Furthermore, the current time domain characteristics include a moving average, a moving variance, a moving range and a change rate average, and the current frequency domain characteristics include a spectrum average, a spectrum standard deviation, a fundamental component ratio and a relative component fundamental component ratio.

[0009] Furthermore, the following contents are also included:

[0010] The time series features of the current curves of all identified residential power facilities and the current curves of the residential power facilities to be identified are extracted respectively, and the extracted time series features are correspondingly integrated into the current feature vector.

[0011] Furthermore, the process of extracting time series features includes the following:

[0012] The STL algorithm is used to decompose the current series into trend component, seasonal component and random component, and the change rate index of the trend component, the periodicity index of the seasonal component and the fluctuation index of the random component are calculated based on the decomposition results.

[0013] Furthermore, the rate of change indicator of the trend component is calculated based on the following formula:

[0014]

[0015] Among them, Tr.rate represents the rate of change index of the trend component, T t-1 and T t They represent the trend components at time (t-1) and time t respectively.

[0016] Furthermore, the periodicity index of seasonal composition is calculated based on the following formula:

[0017]

[0018] Among them, Se.str represents the periodicity index of seasonal components, σ(S t ) and μ(St ) represent the standard deviation and mean of the seasonal components, respectively.

[0019] Furthermore, the volatility index of the stochastic component is calculated based on the following formula:

[0020]

[0021] Among them, Re.vol represents the volatility index of the random component, n represents the number of random components, R i represents the i-th random component.

[0022] In addition, the present invention also provides a classification and identification system for residential power facilities, comprising:

[0023] The first feature extraction module is used to obtain the current curves of all identified residential power facilities, extract the current time domain features and current frequency domain features of each identified residential power facility, so as to construct a current feature vector of each identified residential power facility;

[0024] The second feature extraction module is used to obtain the current curve of the residential power facility to be identified, extract the current time domain characteristics and current frequency domain characteristics of the residential power facility to be identified, so as to construct a current feature vector of the residential power facility to be identified;

[0025] The load classification and identification module is used to calculate the correlation between the current feature vector of the residential power facility to be identified and the current feature vectors of all identified residential power facilities, and take the identified residential power facility with the greatest correlation as the classification and identification result of the residential power facility to be identified.

[0026] In addition, the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0027] In addition, the present invention also provides a computer-readable storage medium for storing a computer program for classifying and identifying residential electrical facilities, wherein the computer program executes the steps of the method described above when running on a computer.

[0028] The present invention has the following beneficial effects:

[0029] The method for classifying and identifying residential power facilities of the present invention first extracts current time domain characteristics and current frequency domain characteristics based on the current curves of all identified residential power facilities and the residential power facilities to be identified, so as to respectively construct current feature vectors of each identified residential power facility and the current feature vectors of the residential power facilities to be identified. Since the time domain characteristics and frequency domain characteristics of the current curves are extracted as current characteristic indicators, the different current characteristics of different residential power facilities can be described in an all-round way, which is beneficial to improving the accuracy of load classification and identification. Moreover, only current data is collected for analysis. Compared with voltage and power, current has the advantage of not being affected by voltage fluctuations, thereby ensuring the accuracy and reliability of the collected data and further improving the accuracy of load classification and identification. Then, the correlation between the current feature vector of the residential power facilities to be identified and the current feature vectors of all identified residential power facilities is calculated. The higher the correlation between the two current feature vectors, the closer their current characteristics are. Therefore, the identified residential power facilities with the largest correlation are used as the classification and identification results of the residential power facilities to be identified. Compared with the existing model classification and identification methods, there is no need to train the model. Only the current feature vectors of all known residential power facilities need to be pre-stored, so as to realize rapid classification and identification of the residential power facilities to be identified when they are newly connected to the power grid, which can well meet the real-time classification and identification needs.

[0030] In addition, the classification and identification system for residential electrical facilities of the present invention also has the above advantages.

[0031] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0033] Figure 1 It is a flow chart of a classification and identification method of residential electrical facilities in a preferred embodiment of the present application;

[0034] Figure 2 It is another flow chart of the classification and identification method of residential power facilities in the preferred embodiment of the present application;

[0035] Figure 3 It is a schematic diagram of the module structure of a classification and identification system for residential electrical facilities according to another embodiment of the present application. DETAILED DESCRIPTION

[0036] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0037] Reference Figure 1 The preferred embodiment of the present application provides a classification and identification method for residential power facilities, including the following contents:

[0038] Step S1: obtaining current curves of all identified residential power facilities, extracting the current time domain characteristics and current frequency domain characteristics of each identified residential power facility, so as to construct a current feature vector of each identified residential power facility;

[0039] Step S2: Obtain the current curve of the residential power facilities to be identified, and extract the current curve of the residential power facilities to be identified. The current time domain characteristics and current frequency domain characteristics of the facilities are used to construct the current feature vector of the residential power facilities to be identified;

[0040] Step S3: Calculate the correlation between the current feature vector of the residential power facility to be identified and the current feature vectors of all identified residential power facilities, and take the identified residential power facility with the greatest correlation as the classification and identification result of the residential power facility to be identified.

[0041] It can be understood that the classification and identification method of residential power facilities in this embodiment first extracts current time domain characteristics and current frequency domain characteristics based on the current curves of all identified residential power facilities and residential power facilities to be identified, so as to respectively construct current feature vectors of each identified residential power facility and the current feature vector of the residential power facilities to be identified. Since the time domain characteristics and frequency domain characteristics of the current curves are extracted as current characteristic indicators, the different current characteristics of different residential power facilities can be described in an all-round way, which is conducive to improving the accuracy of load classification and identification. Moreover, only current data is collected for analysis. Compared with voltage and power, current has the advantage of not being affected by voltage fluctuations, which ensures the accuracy and reliability of the collected data and further improves the accuracy of load classification and identification. Then, the correlation between the current feature vector of the residential power facilities to be identified and the current feature vectors of all identified residential power facilities is calculated. The higher the correlation between the two current feature vectors, the closer their current characteristics are. Therefore, the identified residential power facilities with the largest correlation are used as the classification and identification results of the residential power facilities to be identified. Compared with the existing model classification and identification methods, there is no need to train the model. Only the current feature vectors of all known residential power facilities need to be pre-stored, so as to realize rapid classification and identification of the residential power facilities to be identified when they are newly connected to the power grid, which can well meet the real-time classification and identification needs.

[0042] It can be understood that in step S1, the current curves of all known residential power facilities in the power grid system are collected, and then the current time domain characteristics and current frequency domain characteristics of each known residential power facility are extracted, so that the current feature vector of each known residential power facility can be constructed. Among them, the current time domain characteristics include moving average, moving variance, moving range and change rate average, and the current frequency domain characteristics include spectrum average, spectrum standard deviation, fundamental component proportion and relative component fundamental proportion. Among them, the collected current curve can be a minute-level curve, a fifteen-minute-level curve or an hour-level curve, and the hour-level curve is preferably used.

[0043] Specifically, the moving average value of the current is calculated based on the following formula:

[0044]

[0045] Among them, I mean represents the moving average value of the current, N represents the number of cycles of the moving average, for example, 24, I i It represents the current value of the power facility at the i-th time point. It can be understood that the moving average can characterize the fluctuation of the current curve, quantitatively represent the overall trend of the current of different power facilities during operation, and reflect the basic situation and change trend of the load of different residential power facilities.

[0046] In addition, the moving variance of the current is calculated based on the following formula: Among them, I var It represents the moving variance of the current. The moving range of the current is calculated based on the following formula: diff =max(I N )-min(I N ), where I diff Indicates the moving range of current, max(I N ) and min(I N ) represent the maximum current and the minimum current within the moving average period respectively.

[0047] It can be understood that the moving variance and moving range of current can reflect the degree and range of current fluctuations in the power consumption process of different residential power facilities, can well characterize the volatility and stability of the loads of different power facilities, and are conducive to identifying possible abnormal conditions in the current curve.

[0048] In addition, the average value of the rate of change of the current is calculated based on the following formula: Among them, I aveIt represents the average value of the rate of change of the current, and Δt represents the time interval. It can be understood that the average value of the rate of change of the current can describe the rate of change of the current curve of different residential power facilities, and well characterize the current volatility and change frequency of different power facilities, so as to evaluate the stability of the current change of different power facilities during their operation.

[0049] Then, the current curve is transformed from the time domain to the frequency domain through Fourier transform to obtain the current spectrum, and then the current frequency domain characteristics are extracted based on the current spectrum. Specifically, the average value of the current spectrum is calculated based on the following formula: Among them, X mean represents the average value of the current spectrum within the moving average period, X i represents the i-th frequency value; the spectrum standard deviation is calculated based on the following formula: Among them, X std Represents the standard deviation of the spectrum.

[0050] It can be understood that the spectrum average and spectrum standard deviation can characterize the spectrum characteristics of the current curves of different power facilities, and can reflect the size and distribution of different frequency components in the current spectra of different power facilities, thereby well reflecting the periodic changes in the current of different residential power facilities.

[0051] In addition, the fundamental component ratio is calculated based on the following formula: Among them, X fud Indicates the proportion of fundamental wave component, f c Indicates the fundamental frequency of the power facility; the relative component fundamental frequency ratio is calculated based on the following formula: Among them, X fc It indicates the relative component fundamental wave proportion, 3rdhc and 5thhc indicate the third harmonic and fifth harmonic of the power facilities respectively. It can be understood that the fundamental wave indicates the most important frequency component of the power facility load, the fundamental wave proportion can reflect the basic frequency characteristics of different power facilities, and the relative component fundamental wave proportion takes into account the influence of harmonics and clutter, and can describe the spectrum properties of different residential power facilities.

[0052] The constructed current characteristic vector can be expressed as: S = [I mean I var I diff I ave X mean X std X fud X fc ], S represents the current characteristic vector.

[0053] It can be understood that the present invention adopts the above four time domain characteristics and four frequency domain characteristics as current characteristic indicators, comprehensively considers the stability, volatility, spectrum characteristics and other aspects of the current of different residential power facilities during operation, and can comprehensively describe the current characteristics and current change process of different residential power facilities, which is conducive to improving the accuracy of load classification and identification. In addition, only current data is collected for analysis. Compared with voltage and power, current has the advantage of not being affected by voltage fluctuations, which ensures the accuracy and reliability of the collected data and further improves the accuracy of load classification and identification.

[0054] Optionally, in step S1, after collecting the current curve, the current data can also be filtered and normalized. For example, first use VSCode software to perform a visual analysis of the current data to determine whether there is missing data and abnormal data in the data. For missing data, comprehensively consider the data of the previous moment, the next moment, the previous day, and the next day, and then use the weight coefficient to perform weighted summing and filling of the data. For abnormal data, horizontal and vertical processing can be used for correction; then, use Gaussian filtering to remove noise and perform weighted averaging of the values ​​of each point on the current curve. For each numerical point, use a Gaussian function to perform weighted averaging of other numerical points in its neighborhood and then perform correction to obtain the load value after noise removal; finally, use the maximum and minimum normalization algorithm to normalize the data after noise removal. Among them, Gaussian filtering is an algorithm for smoothing one-dimensional data curves, and its expression is: Among them, μ represents the mean and σ represents the standard deviation.

[0055] It can be understood that in step S2, the current curve of the residential power facilities to be identified is collected, and then the current time domain characteristics and current frequency domain characteristics are extracted using the feature extraction method as in the above step S1, and the current feature vector of the residential power facilities to be identified is also constructed, wherein the specific process is the same as step S1 and will not be repeated here.

[0056] It can be understood that in step S3, the cosine similarity algorithm or the Pearson correlation coefficient algorithm is used to calculate the correlation between the current feature vector of the residential power facility to be identified and the current feature vectors of all identified residential power facilities. The higher the correlation, the higher the similarity of the two current curves, and the closer the current characteristics of the two are. Therefore, the identified residential power facility with the greatest correlation is used as the classification and identification result of the residential power facility to be identified, which can quickly and accurately realize the classification and identification of the residential power facilities to be identified. Compared with the existing model classification and identification method, there is no need to train the model, and only the current feature vectors of all known residential power facilities need to be pre-stored, so as to realize the rapid classification and identification of the residential power facilities to be identified when they are newly connected to the power grid, which can well meet the real-time classification and identification needs.

[0057] Alternatively, if Figure 2 As shown, the classification and identification method of residential power facilities further includes the following contents after constructing the current feature vectors of all identified residential power facilities and the current feature vectors of the residential power facilities to be identified:

[0058] Step S20: extracting the time series features of the current curves of all identified residential power facilities and the current curves of the residential power facilities to be identified respectively, and correspondingly integrating the extracted time series features into the current feature vector.

[0059] The process of extracting time series features includes the following:

[0060] The STL algorithm is used to decompose the current series into trend component, seasonal component and random component, and the change rate index of the trend component, the periodicity index of the seasonal component and the fluctuation index of the random component are calculated based on the decomposition results.

[0061] Specifically, the STL (Seasonal-Trend decomposition using Loess) algorithm is first used to decompose the current sequence (i.e., the current curve) into trend components, seasonal components, and random components. The specific decomposition process belongs to the prior art and will not be repeated here. Then, based on the decomposition results, the change rate index of the trend component, the periodicity index of the seasonal component, and the fluctuation index of the random component are calculated. Specifically, the change rate index of the trend component is calculated based on the following formula:

[0062]

[0063] Among them, Tr.rate represents the rate of change index of the trend component, T t-1 and T t They represent the trend components at time (t-1) and time t respectively.

[0064] It can be understood that the trend component is often relatively stable and continuous, while the rate of change of the trend reflects the rate of long-term trend change in the current curves of different power facilities, and describes the evolution of the current consumption pattern. By analyzing the trend change rate of the current load, the system can identify the long-term growth trend or decline trend of the load, and at the same time eliminate the impact of short-term noise, thereby effectively removing the impact of short-term fluctuations from the data, focusing more on the long-term load change trend, and helping to improve the accuracy of classification and identification.

[0065] In addition, the seasonality index is calculated based on the following formula:

[0066]

[0067] Among them, Se.str represents the periodicity index of seasonal components, σ(St ) and μ(S t ) represent the standard deviation and mean of the seasonal components, respectively.

[0068] It can be understood that seasonal changes are usually manifested as an increase or decrease in current load in a specific time period (such as summer or winter), and the periodic indicators of the seasonal component can reflect the periodic fluctuation pattern of the current curve of different power facilities. This fluctuation is usually caused by seasonal factors and is conducive to revealing the typical periodic patterns of different residential power facilities.

[0069] In addition, the volatility indicator of the stochastic component is calculated based on the following formula:

[0070]

[0071] Among them, Re.vol represents the volatility index of the random component, n represents the number of random components, R i represents the i-th random component.

[0072] It can be understood that the random component is mainly caused by some unforeseen random events or short-term interferences, and the fluctuation degree of the random component reflects the noise and abnormal fluctuations in the current curves of different power facilities, which helps to identify abnormal conditions in the current load and further eliminate the impact of these abnormal values ​​on the overall load identification.

[0073] Next, the calculated rate of change index of the trend component, the periodic index of the seasonal component and the fluctuation index of the random component are respectively fused into the current feature vector as a feature. The fused current feature vector includes four time domain features, four frequency domain features and three time series features. Subsequently, correlation calculation is performed based on the fused feature vector.

[0074] It can be understood that the present invention also decomposes the current curve into trend component, seasonal component and random component, and extracts the change rate index of the trend component, the periodicity index of the seasonal component and the fluctuation index of the random component, and further extracts the characteristics of the current data from multiple levels, which can capture the long-term changes, periodic fluctuations and random fluctuations of the current load respectively, and can provide more fine-grained information, which is conducive to more comprehensive reflection of the current load characteristics of different residential power facilities. The comprehensive use of these characteristics can improve the accuracy of load identification, especially when facing complex load change patterns, and can significantly improve the identification accuracy, stability and robustness.

[0075] In addition, if Figure 3 As shown, another embodiment of the present invention further provides a classification and identification system for residential power facilities, preferably using the classification and identification method for residential power facilities as described above, the system comprising:

[0076] The first feature extraction module is used to obtain the current curves of all identified residential power facilities, extract the current time domain features and current frequency domain features of each identified residential power facility, so as to construct a current feature vector of each identified residential power facility;

[0077] The second feature extraction module is used to obtain the current curve of the residential power facility to be identified, extract the current time domain characteristics and current frequency domain characteristics of the residential power facility to be identified, so as to construct a current feature vector of the residential power facility to be identified;

[0078] The load classification and identification module is used to calculate the correlation between the current feature vector of the residential power facility to be identified and the current feature vectors of all identified residential power facilities, and take the identified residential power facility with the greatest correlation as the classification and identification result of the residential power facility to be identified.

[0079] It can be understood that the classification and identification system for residential power facilities in the present embodiment first extracts the time domain characteristics and the frequency domain characteristics of the current based on the current curves of all identified residential power facilities and the residential power facilities to be identified, so as to respectively construct the current feature vectors of each identified residential power facility and the current feature vectors of the residential power facilities to be identified. Since the time domain characteristics and the frequency domain characteristics of the current curves are extracted as current characteristic indicators, the different current characteristics of different residential power facilities can be described in all aspects, which is conducive to improving the accuracy of load classification and identification. Moreover, only current data is collected for analysis. Compared with voltage and power, current has the advantage of not being affected by voltage fluctuations, which ensures the accuracy and reliability of the collected data and further improves the accuracy of load classification and identification. Then, the correlation between the current feature vector of the residential power facilities to be identified and the current feature vectors of all identified residential power facilities is calculated. The higher the correlation between the two current feature vectors, the closer their current characteristics are. Therefore, the identified residential power facilities with the largest correlation are used as the classification and identification results of the residential power facilities to be identified. Compared with the existing model classification and identification methods, there is no need to train the model. Only the current feature vectors of all known residential power facilities need to be pre-stored, so as to realize rapid classification and identification of the residential power facilities to be identified when they are newly connected to the power grid, which can well meet the real-time classification and identification needs.

[0080] In addition, another embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0081] In addition, another embodiment of the present invention further provides a computer-readable storage medium for storing a computer program for classifying and identifying residential electrical facilities, wherein the computer program executes the steps of the method described above when running on a computer.

[0082] The general form of computer readable storage media includes: floppy disk, flexible disk, hard disk, magnetic tape, any other magnetic medium, CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with a pattern of holes, random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), flash erasable programmable read-only memory (FLASH-EPROM), any other memory chip or cartridge, or any other medium that can be read by a computer. The instructions can further be transmitted or received by a transmission medium. The term transmission medium can include any tangible or intangible medium that can be used to store, encode or carry instructions for execution by a machine, and includes digital or analog communication signals or other intangible media that facilitate the communication of the above instructions. Transmission media include coaxial cables, copper wires and optical fibers, which include the wires of a bus used to transmit a computer data signal.

[0083] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0084] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0086] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented processing. management, so that in the computer or Instructions executed on other programmable devices are provided to implement Figure 1 indivual Process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0087] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0088] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A classification and identification method for residential power facilities, characterized in that: Includes the following: Obtaining current curves of all identified residential power facilities, extracting current time domain features and current frequency domain features of each identified residential power facility, so as to construct a current feature vector of each identified residential power facility; Obtaining a current curve of the residential power facility to be identified, extracting the current time domain characteristics and current frequency domain characteristics of the residential power facility to be identified, so as to construct a current feature vector of the residential power facility to be identified; The correlation between the current feature vector of the residential power facility to be identified and the current feature vectors of all identified residential power facilities is calculated, and the identified residential power facility with the greatest correlation is taken as the classification and identification result of the residential power facility to be identified.

2. The method for classifying and identifying residential electrical facilities according to claim 1, characterized in that: The current time domain characteristics include moving average, moving variance, moving range and change rate average, and the current frequency domain characteristics include spectrum average, spectrum standard deviation, fundamental component proportion and relative component fundamental component proportion.

3. The classification and identification method of residential electrical facilities according to claim 1, characterized in that: Also included are the following: The time series features of the current curves of all identified residential power facilities and the current curves of the residential power facilities to be identified are extracted respectively, and the extracted time series features are correspondingly integrated into the current feature vector.

4. The method for classifying and identifying residential electrical facilities according to claim 3, characterized in that: The process of extracting time series features includes the following: The STL algorithm is used to decompose the current series into trend component, seasonal component and random component, and the change rate index of the trend component, the periodicity index of the seasonal component and the fluctuation index of the random component are calculated based on the decomposition results.

5. The method for classifying and identifying residential electrical facilities according to claim 4, characterized in that: The rate of change indicator of the trend component is calculated based on the following formula: Among them, Tr.rate represents the rate of change of the trend component, T t-1 and T t They represent the trend components at time (t-1) and time t respectively.

6. The method for classifying and identifying residential electrical facilities according to claim 4, characterized in that: The seasonality index is calculated based on the following formula: Among them, Se.str represents the periodicity index of seasonal components, σ(S t ) and μ(S t ) represent the standard deviation and mean of the seasonal components, respectively.

7. The method for classifying and identifying residential electrical facilities according to claim 4, characterized in that: The volatility indicator of the stochastic component is calculated based on the following formula: Among them, Re.vol represents the volatility index of the random component, n represents the number of random components, R i represents the i-th random component.

8. A classification and identification system for residential electrical facilities, characterized in that: include: The first feature extraction module is used to obtain the current curves of all identified residential power facilities, extract the current time domain features and current frequency domain features of each identified residential power facility, so as to construct a current feature vector of each identified residential power facility; The second feature extraction module is used to obtain the current curve of the residential power facility to be identified, extract the current time domain characteristics and current frequency domain characteristics of the residential power facility to be identified, so as to construct a current feature vector of the residential power facility to be identified; The load classification and identification module is used to calculate the correlation between the current feature vector of the residential power facility to be identified and the current feature vectors of all identified residential power facilities, and take the identified residential power facility with the greatest correlation as the classification and identification result of the residential power facility to be identified.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the method according to any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium for storing a computer program for classifying and identifying residential electrical facilities, characterized in that: When the computer program is run on a computer, the steps of the method according to any one of claims 1 to 7 are executed.

Images