Non-intrusive load identification method and device, terminal equipment and storage medium

Abstract

The invention discloses a non-intrusive load identification method and device, terminal equipment and a storage medium. The method comprises the following steps: acquiring a voltage oscillogram of a main loop of a power distribution network in a period and a current effective value of each cycle wave in the period; when it is determined that new load starting exists in the period according to the current effective value, the starting moment of the load is determined according to the voltage oscillogram, and the effective waveform of the load during operation is extracted from the voltage oscillogram according to the starting moment; and extracting load characteristics from the effective waveform and inputting the load characteristics into a preset load identification model, so that the load identification model determines the type of a load started in the period according to the load characteristics. Therefore, the load characteristics of the newly started load during operation are extracted before load identification, so that the load identification model performs load identification by considering the load characteristics of the time-space characteristics, other irrelevant and complicated electrical quantity data do not need to be processed, and the accuracy of the identification result of the load identification model is improved.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a non-intrusive load identification method, device, terminal equipment and storage medium. Background Art

[0002] Load identification refers to the process of identifying and distinguishing the load in the power system or equipment through certain technical means. Load identification is mainly used to determine the basic energy consumption level and load demand of the system or equipment under normal operating conditions, so as to provide a basis for monitoring and optimizing the system operating status. Therefore, load identification is widely used in power systems. For example, in the planning stage of the power system, load identification can help determine the changing trend of future power demand and provide a basis for the expansion and transformation of the power grid; in the dispatching stage of the power system, load identification can help dispatchers better understand the current load status of the system and formulate reasonable dispatching plans; in the operation stage of the power system, load identification can help monitor and diagnose the operating status of the equipment, and promptly discover and deal with potential faults and problems. In short, load identification is an indispensable part of the operation and management of the power system, and it is of great significance to improve the operating efficiency, stability and economy of the power system.

[0003] At present, there are two main types of load identification methods: intrusive load identification and non-intrusive load identification. Among them, although the intrusive load identification method has relatively accurate identification results, it is not very popular due to the high cost of circuit modification and installation of a large number of sensors. Non-intrusive load identification refers to the use of intelligent algorithms to analyze and identify the type, working status and energy consumption of each electrical equipment without intruding into the user's electrical equipment, only by monitoring the electrical quantity data. At present, load identification is mainly carried out through algorithms such as invisible Markov models and support vector machines, but these algorithms can only process simple electrical quantity data. Due to the increasingly complex structure of modern power grids, the electrical quantity data of various power grid loads are relatively complex, which makes the error rate of the identification results of the existing non-intrusive load identification models relatively high. Therefore, how to improve the accuracy of load identification has become a problem that needs to be solved urgently. Summary of the invention

[0004] The embodiments of the present invention provide a non-intrusive load identification method, apparatus, terminal device and storage medium, which extracts the load characteristics of a newly started load during operation before load identification, so that the load identification model only needs to perform load identification based on the load characteristics without processing other irrelevant and complex electrical quantity data, thereby improving the accuracy of the identification results of the load identification model.

[0005] An embodiment of the present invention provides a non-intrusive load identification method, comprising:

[0006] Obtaining a voltage waveform diagram of the main circuit of the distribution network within a cycle and an effective current value of each cycle within the cycle;

[0007] Performing load startup detection according to the effective value of the current and generating a detection result;

[0008] When it is determined according to the detection result that a new load starts within the period, the start time of the load is determined according to the voltage waveform diagram, and according to the start time, the effective waveform of the load during operation is extracted from the voltage waveform diagram;

[0009] The load characteristics are extracted from the effective waveform and input into a preset load identification model, so that the load identification model determines the type of load started in the cycle according to the load characteristics.

[0010] Further, a load startup detection is performed according to the effective value of the current, and a detection result is generated, including:

[0011] Get a periodic sliding window of preset length;

[0012] Extracting each frequency set within the period one by one according to the period sliding window;

[0013] When each cycle set is extracted, determining whether the effective value of the target current in the current cycle sliding window fluctuates;

[0014] If so, the number of current fluctuations in the cycle is increased by one;

[0015] When the number of changes exceeds a preset number threshold, a detection result is generated for indicating that a new load is started within the cycle.

[0016] Further, the determining whether the effective value of the target current in the current period sliding window fluctuates includes:

[0017] According to the cycle sequence of the target current effective value, calculating the absolute difference between the target current effective values ​​of the first two cycles and the target current effective values ​​of the last two cycles;

[0018] When the absolute difference is not less than a preset absolute difference threshold, determining that the effective value of the target current in the current period sliding window fluctuates;

[0019] When the absolute difference is less than a preset absolute difference threshold, it is determined that the effective value of the target current in the current period sliding window does not fluctuate.

[0020] Further, the determining whether the effective value of the target current in the current period sliding window fluctuates includes:

[0021] According to the cycle sequence of the target current effective value, calculating the change rate of the target current effective value of the first two cycles and the target current effective value of the last two cycles;

[0022] When the change rate is not less than a preset change rate threshold, determining that the effective value of the target current in the current period sliding window fluctuates;

[0023] When the change rate is less than a preset change rate threshold, it is determined that the effective value of the target current in the current period sliding window does not fluctuate.

[0024] Furthermore, extracting the effective waveform of the load during operation from the voltage waveform diagram according to the startup time includes:

[0025] Extracting a first steady-state waveform before the start-up moment and a second steady-state waveform after the start-up moment from the voltage waveform diagram;

[0026] The first steady-state waveform is subtracted from the second steady-state waveform to separate the effective waveform during the load operation.

[0027] Furthermore, the load characteristics include: current characteristic value, power, fundamental wave, and harmonics;

[0028] The step of extracting the load characteristic from the effective waveform comprises:

[0029] According to the effective waveform, extracting a corresponding current characteristic value from a number of the current effective values;

[0030] Calculating the power according to the effective waveform and the current characteristic value;

[0031] Perform Fourier transform on the effective waveform to extract the fundamental wave and the harmonics.

[0032] Furthermore, the non-intrusive load identification method described in any one of the above embodiments of the invention further includes:

[0033] Obtain several historical valid waveforms of various loads under operating conditions;

[0034] Extracting load characteristic samples from the plurality of historical valid waveforms, and assigning a corresponding load type label to each load characteristic sample according to the load to which each load characteristic sample belongs, thereby constructing a training sample set;

[0035] Construct an initial model based on BP neural network;

[0036] Taking the load characteristic samples of the training sample set as input and the load type labels of the training sample set as output, the initial model is trained, and during the training process, a preset loss function is used to evaluate the output result of the initial model;

[0037] When the evaluation result generated by the loss function converges and is minimized, it is determined that the initial model has completed training, and the trained initial model is used as the load identification model.

[0038] Another embodiment of the present invention provides a non-intrusive load identification device, comprising:

[0039] A data acquisition module is used to obtain a voltage waveform diagram of the main circuit of the distribution network within a cycle and an effective current value of each cycle within the cycle;

[0040] A start detection module, used for performing load start detection according to the effective value of the current and generating a detection result;

[0041] A waveform acquisition module, for determining the start-up time of the load according to the voltage waveform diagram when it is determined according to the detection result that a new load starts within the period, and extracting the effective waveform of the load when running from the voltage waveform diagram according to the start-up time;

[0042] The load identification module is used to extract the load characteristics from the effective waveform and input the characteristics into a preset load identification model, so that the load identification model determines the load type started in the cycle according to the load characteristics.

[0043] Another embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, a non-intrusive load identification method as described in any one of the embodiments is implemented.

[0044] Another embodiment of the present invention provides a storage medium, characterized in that the storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute a non-intrusive load identification method as described in any one of the above embodiments.

[0045] The following beneficial effects are achieved by implementing the present invention:

[0046] The present invention discloses a non-intrusive load identification method, device, terminal equipment and storage medium. The method first determines whether there is a new load started in a cycle by the effective value of the current in the main circuit of the distribution network in the cycle before performing load identification. When it is determined that there is a new load started in the cycle, the effective waveform is extracted from the voltage waveform diagram of the cycle, and the load characteristics are extracted from the effective waveform and input into a preset load identification model, so that the load identification model determines the type of load started in the cycle according to the load characteristics. Therefore, the present invention extracts the load characteristics of the newly started load when it is running before performing load identification, so that the load identification model only needs to perform load identification according to the load characteristics, without processing other irrelevant and complex electrical quantity data, so as to improve the accuracy of the identification result of the load identification model. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flowchart of a non-intrusive load identification method provided by an embodiment of the present invention.

[0048] Figure 2 It is a structural schematic diagram of a non-intrusive load identification device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0051] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0052] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0053] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0054] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0055] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0056] See also Figure 1 , is a flow chart of a non-intrusive load identification method provided by an embodiment of the present invention, comprising:

[0057] S1. Obtaining a voltage waveform diagram of the main circuit of the distribution network within a cycle and the effective value of the current of each cycle within the cycle;

[0058] In a preferred embodiment of the present invention, the cycle refers to a complete cycle of alternating current (AC). In alternating current, the direction and magnitude of the current will change over time, and this change is periodic. Specifically, a cycle refers to the time period from a positive peak value of the current, through a negative peak value, and then back to the original positive peak value. Furthermore, in a cycle, the current or voltage will change periodically in the form of a sine wave (or other waveform), and a complete waveform change is a cycle. The effective value (RMS, Root Mean Square) of the current is an equivalent DC value that measures the size of the alternating current, which takes into account the fluctuation of the current over time.

[0059] S2. Perform load startup detection according to the effective value of the current and generate a detection result;

[0060] Preferably, performing load startup detection according to the effective value of the current and generating a detection result includes:

[0061] S21, obtaining a periodic sliding window of a preset length;

[0062] S22, extracting each frequency set within the period one by one according to the period sliding window;

[0063] S23, when each cycle set is extracted, determining whether the effective value of the target current in the current cycle sliding window fluctuates;

[0064] Preferably, the determining whether the effective value of the target current in the current period sliding window fluctuates includes:

[0065] S2311, calculating the absolute difference between the target current effective values ​​of the first two cycles and the target current effective values ​​of the last two cycles according to the cycle sequence of the target current effective value;

[0066] S2312: when the absolute difference is not less than a preset absolute difference threshold, determining that the effective value of the target current in the current period sliding window fluctuates;

[0067] S2313: When the absolute difference is less than a preset absolute difference threshold, determine that the effective value of the target current in the current period sliding window does not fluctuate.

[0068] Preferably, the determining whether the effective value of the target current in the current period sliding window fluctuates includes:

[0069] S2321, calculating the change rate of the target current effective values ​​of the first two cycles and the target current effective values ​​of the last two cycles according to the cycle sequence of the target current effective values;

[0070] S2322, when the change rate is not less than a preset change rate threshold, determining that the effective value of the target current in the current period sliding window fluctuates;

[0071] S2323: When the change rate is less than a preset change rate threshold, determine that the effective value of the target current in the current period sliding window does not fluctuate.

[0072] S24, if yes, then the number of current fluctuations in the cycle is increased by one;

[0073] S25. When the number of changes exceeds a preset number threshold, generate a detection result for indicating that a new load is started within the cycle.

[0074] In a preferred embodiment of the present invention, a periodic sliding window with a length of θ cycles is set to detect whether the effective current value changes in the previous and next cycles in the sliding window. If the change exceeds the threshold value σ, the count is increased by 1. In order to avoid the influence of current noise, if the number of continuous changes in the effective current value exceeds the number threshold m during the movement of the sliding window, it is qualitatively determined that the load is started;

[0075] Specifically, the first two cycles and the last two cycles of the period sliding window θ cycles are taken to calculate the target current effective value, and it is determined whether the change of the target current effective value exceeds the fluctuation threshold. There are usually two thresholds, one is the change rate threshold and the other is the absolute difference threshold. In actual application scenarios, any calculation method can be selected according to actual needs to determine whether the target current effective value fluctuates.

[0076] S3, when it is determined according to the detection result that a new load starts within the cycle, determining the start time of the load according to the voltage waveform diagram, and extracting the effective waveform of the load when running from the voltage waveform diagram according to the start time;

[0077] Preferably, extracting the effective waveform of the load during operation from the voltage waveform diagram according to the startup time includes:

[0078] S31, extracting a first steady-state waveform before the startup moment and a second steady-state waveform after the startup moment from the voltage waveform diagram;

[0079] S32: Subtract the first steady-state waveform from the second steady-state waveform to separate the effective waveform during load operation.

[0080] In a preferred embodiment of the present invention, the zero-crossing point is determined based on the voltage waveform, and the zero-crossing point is the specific time t when the load starts; the zero-crossing point is determined as follows: if the values ​​of five consecutive points in the waveform decrease, and the first two points are greater than zero, and the last two points are less than zero, then the middle point is determined to be the lower zero-crossing point; if the values ​​of five consecutive points in the waveform increase, and the first two points are less than zero, and the last two points are greater than zero, then the middle point is determined to be the upper zero-crossing point.

[0081] Specifically, in this embodiment, the steady-state region in the voltage waveform is first identified, which is usually the stable part before and after the load startup time t. Taking t as the dividing point, the steady-state waveforms before and after t are respectively intercepted. The steady-state waveform after t is subtracted from the steady-state waveform before t to obtain the voltage change waveform caused by the load operation, that is, the effective waveform.

[0082] S4. Extracting load characteristics from the effective waveform and inputting them into a preset load identification model, so that the load identification model determines the type of load started within the cycle according to the load characteristics.

[0083] Preferably, the load characteristics include: current characteristic value, power, fundamental wave, and harmonics;

[0084] The step of extracting the load characteristic from the effective waveform comprises:

[0085] S41, extracting a corresponding current characteristic value from a number of current effective values ​​according to the effective waveform;

[0086] S42, calculating the power according to the effective waveform and the current characteristic value;

[0087] S43, performing Fourier transform on the effective waveform to extract the fundamental wave and the harmonics.

[0088] Preferably, the non-intrusive load identification method described in any one of the above embodiments of the invention further includes:

[0089] S01. Obtain several historical valid waveforms of various loads in operation state;

[0090] S02, extracting load characteristic samples from the plurality of historical valid waveforms, and assigning each load characteristic sample a corresponding load type label according to the load to which each load characteristic sample belongs, thereby constructing a training sample set;

[0091] S03, constructing an initial model based on BP neural network;

[0092] S04, taking the load characteristic samples of the training sample set as input and the load type labels of the training sample set as output, training the initial model, and during the training process, using a preset loss function to evaluate the output result of the initial model;

[0093] S05. When the evaluation result generated by the loss function converges and is minimized, it is determined that the initial model has completed training, and the trained initial model is used as the load identification model.

[0094] In a preferred embodiment of the present invention, based on the historically collected startup and operating status frequency data of various electrical appliances (such as kettles, televisions, battery vehicles, etc.), the effective current value, power, fundamental wave, harmonics and other features are separated and recorded as X. The load type corresponding to the feature X is regarded as a classification label, recorded as Y. The feature X and the label Y are used to train a neural network for load classification. The neural network uses a two-layer BP neural network, recorded as C. The loss function of the training process is as follows:

[0095]

[0096] Where n is the number of data, Yi is the load type corresponding to the i-th data, and Xi is the characteristic value corresponding to the i-th data.

[0097] This embodiment provides a non-intrusive load identification method, which first determines whether there is a new load started in a cycle by the effective value of the current in the main circuit of the distribution network before load identification. When it is determined that there is a new load started in the cycle, the effective waveform is extracted from the voltage waveform diagram of the cycle, and the load characteristics are extracted from the effective waveform and input into a preset load identification model, so that the load identification model determines the type of load started in the cycle according to the load characteristics. Therefore, the present invention extracts the load characteristics of the newly started load when it is running before load identification, so that the load identification model only needs to perform load identification according to the load characteristics, without processing other irrelevant and complex electrical quantity data, so as to improve the accuracy of the identification result of the load identification model.

[0098] See also Figure 2 , is a schematic diagram of a non-intrusive load identification device provided by an embodiment of the present invention, comprising:

[0099] A data acquisition module is used to obtain a voltage waveform diagram of the main circuit of the distribution network within a cycle and an effective current value of each cycle within the cycle;

[0100] A start detection module, used for performing load start detection according to the effective value of the current and generating a detection result;

[0101] A waveform acquisition module, for determining the start-up time of the load according to the voltage waveform diagram when it is determined according to the detection result that a new load starts within the period, and extracting the effective waveform of the load when running from the voltage waveform diagram according to the start-up time;

[0102] The load identification module is used to extract the load characteristics from the effective waveform and input the characteristics into a preset load identification model, so that the load identification model determines the load type started in the cycle according to the load characteristics.

[0103] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0104] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0105] Another preferred embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, a non-intrusive load identification method as described in any one of the above embodiments is implemented.

[0106] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0107] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0108] The memory can be used to store the computer program, and the processor realizes various functions of the terminal device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med i aCard, SMC), a secure digital (Secure Digital, SD) card, a flash card (F l ash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0109] Another preferred embodiment of the present invention provides a storage medium, the storage medium is a computer-readable storage medium, the computer program is stored in the computer-readable storage medium, and the computer program, when executed by the processor, can implement the steps of the above-mentioned various method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0110] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A non-intrusive load identification method, characterized in that: include: Obtaining a voltage waveform diagram of the main circuit of the distribution network within a cycle and an effective current value of each cycle within the cycle; Performing load startup detection according to the effective value of the current and generating a detection result; When it is determined according to the detection result that a new load starts within the period, the start time of the load is determined according to the voltage waveform diagram, and according to the start time, the effective waveform of the load during operation is extracted from the voltage waveform diagram; The load characteristics are extracted from the effective waveform and input into a preset load identification model, so that the load identification model determines the type of load started in the cycle according to the load characteristics.

2. A non-intrusive load identification method as claimed in claim 1, characterized in that: The load startup detection is performed according to the effective value of the current, and a detection result is generated, including: Get a periodic sliding window of preset length; Extracting each frequency set within the period one by one according to the period sliding window; When each cycle set is extracted, determining whether the effective value of the target current in the current cycle sliding window fluctuates; If so, the number of current fluctuations in the cycle is increased by one; When the number of changes exceeds a preset number threshold, a detection result is generated for indicating that a new load is started within the period.

3. A non-intrusive load identification method as claimed in claim 2, characterized in that: The determining whether the effective value of the target current in the current period sliding window fluctuates includes: According to the cycle sequence of the target current effective value, calculating the absolute difference between the target current effective values ​​of the first two cycles and the target current effective values ​​of the last two cycles; When the absolute difference is not less than a preset absolute difference threshold, determining that the effective value of the target current in the current period sliding window fluctuates; When the absolute difference is less than a preset absolute difference threshold, it is determined that the effective value of the target current in the current period sliding window does not fluctuate.

4. A non-intrusive load identification method as claimed in claim 2, characterized in that: The determining whether the effective value of the target current in the current period sliding window fluctuates includes: According to the cycle sequence of the target current effective value, calculating the change rate of the target current effective value of the first two cycles and the target current effective value of the last two cycles; When the change rate is not less than a preset change rate threshold, determining that the effective value of the target current in the current period sliding window fluctuates; When the change rate is less than a preset change rate threshold, it is determined that the effective value of the target current in the current period sliding window does not fluctuate.

5. A non-intrusive load identification method as claimed in claim 1, characterized in that: The step of extracting the effective waveform of the load during operation from the voltage waveform diagram according to the startup time includes: Extracting a first steady-state waveform before the start-up moment and a second steady-state waveform after the start-up moment from the voltage waveform diagram; The first steady-state waveform is subtracted from the second steady-state waveform to separate the effective waveform during the load operation.

6. A non-intrusive load identification method as claimed in claim 1, characterized in that: The load characteristics include: current characteristic value, power, fundamental wave, and harmonics; The step of extracting the load characteristic from the effective waveform comprises: According to the effective waveform, extracting a corresponding current characteristic value from a number of the current effective values; Calculating the power according to the effective waveform and the current characteristic value; Perform Fourier transform on the effective waveform to extract the fundamental wave and the harmonics.

7. A non-intrusive load identification method as claimed in claim 1, characterized in that: Also includes: Obtain several historical valid waveforms of various loads under operating conditions; Extracting load characteristic samples from the plurality of historical valid waveforms, and assigning a corresponding load type label to each load characteristic sample according to the load to which each load characteristic sample belongs, thereby constructing a training sample set; Construct an initial model based on BP neural network; Taking the load characteristic samples of the training sample set as input and the load type labels of the training sample set as output, the initial model is trained, and during the training process, a preset loss function is used to evaluate the output result of the initial model; When the evaluation result generated by the loss function converges and is minimized, it is determined that the initial model has completed training, and the trained initial model is used as the load identification model.

8. A non-intrusive load identification device, characterized in that: include: A data acquisition module is used to obtain a voltage waveform diagram of the main circuit of the distribution network within a cycle and an effective current value of each cycle within the cycle; A start detection module, used for performing load start detection according to the effective value of the current and generating a detection result; A waveform acquisition module, for determining the start-up time of the load according to the voltage waveform diagram when it is determined according to the detection result that a new load starts within the period, and extracting the effective waveform of the load when running from the voltage waveform diagram according to the start-up time; The load identification module is used to extract the load characteristics from the effective waveform and input the characteristics into a preset load identification model, so that the load identification model determines the load type started in the cycle according to the load characteristics.

9. A terminal device, characterized in that: The invention comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, a non-intrusive load identification method as claimed in any one of claims 1 to 7 is implemented.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute a non-intrusive load identification method as described in any one of claims 1 to 7.

Images