User load identification method, fusion terminal, device, equipment and medium
Through sparse encoding and integer planning methods, the decomposition problem of user comprehensive electricity data is converted into integer planning problem. Combined with Monte Carlo algorithm and fast online dictionary update, the problem of low load recognition efficiency in traditional methods is solved, achieving efficient and accurate user load identification.
Patent Information
- Application Number
- CN202411058686.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-08
Smart Images

Figure CN118916638A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power systems, and in particular relates to a user load identification method, a fusion terminal, a device, equipment and a medium. Background Art
[0002] Building a new power system and promoting clean and low-carbon energy transformation require the deep integration of digital technology and physical power grids, and the construction of an IoT perception system covering the entire process of power collection, transmission, storage, and use. Effective demand-side management can not only help the power grid to enhance the operating efficiency of the power grid, but also relieve energy pressure and improve energy utilization efficiency. With the advancement of demand-side management, load identification is an important factor in realizing intelligent management on the demand side. Through load monitoring, the actual energy consumption level of various loads in users can be understood, and the scientific collection and management of energy efficiency data can be realized. Traditional load identification methods adopt an intrusive method, with high hardware costs, complex installation, maintenance and management, and low user acceptance.
[0003] In order to realize user-side load identification, non-intrusive load identification technology has become the mainstream. This method collects electrical signals at the power input, analyzes the types of loads in the user, and realizes load identification, thereby reducing the economic cost of equipment and the difficulty of installation and maintenance. However, the existing non-intrusive load identification technology still faces many challenges in user load identification. For example, in the load identification process, the user's comprehensive electricity consumption data is decomposed into independent loads, which requires the use of complex optimization algorithms, resulting in inefficient load identification. Summary of the invention
[0004] The purpose of the present invention is to provide a user load identification method, a fusion terminal, a device, a equipment and a medium to solve the problem that the traditional optimization algorithm in the prior art is relatively complex, resulting in insufficient efficiency in load identification.
[0005] In order to achieve the above object, the present invention adopts the following technical solution:
[0006] In a first aspect, the present invention provides a method for identifying user load, comprising the following steps:
[0007] Acquire a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding;
[0008] Obtaining comprehensive electricity consumption data of users; using a load identification method based on integer programming, based on the user load feature library, converting the decomposition problem of the comprehensive electricity consumption data of users into an integer programming problem;
[0009] Solve the integer programming problem and identify the independent loads included in the user's comprehensive electricity consumption data.
[0010] Furthermore, the user load feature library is characterized based on sparse coding, including:
[0011] Assume that a user load characteristic library is: in, is the feature of the i-th load in the user load feature library, the feature dimension is N×1, and there are M types of loads in the user load feature library;
[0012] Assume that the over-complete dictionary used to characterize the user load feature library is Contains P groups of basis vectors, and P>>N;
[0013] The user load feature library based on sparse coding is expressed as: L≈DX; where, It is the sparse matrix of user load feature library L.
[0014] Furthermore, a load identification method based on integer programming is adopted, and based on the user load feature library, the decomposition problem of the user's comprehensive power consumption data is converted into an integer programming problem, including:
[0015] Construct the first problem model:
[0016]
[0017] stθ i =0 or 1
[0018] in, is the sparse representation of the user load feature library L under the dictionary D; θ=[θ1,θ2,...θ M ] T is the polynomial coefficient, which is used to characterize each load operation state; X Y is the sparse representation of Y under dictionary D, Y = DX Y , The comprehensive load characteristics of the user's comprehensive electricity consumption data are compared with the individual load characteristics in the user load characteristic library. i have the same characteristics;
[0019] The first problem model is expressed as the second problem model:
[0020] minθ T X T Xθ-2X Y T Xθ+X Y T X Y
[0021] stθ i =0 or 1
[0022] Among them, θ i=0 means that the i-th load is not a component of the current user's comprehensive electricity consumption data; θ i =1 indicates that the i-th load is a component of the current user's comprehensive electricity consumption data.
[0023] Furthermore, the integer programming problem is solved to identify the independent loads contained in the user's comprehensive electricity consumption data, including:
[0024] Let Q = X T X, C = -2X Y T X; due to θ i =0 / 1, then we know that θ i =θ i 2 ;
[0025] Let Q = Q + diag (C), then the second problem model is further expressed as the third problem model:
[0026] minθ T Qθ
[0027] stθ i =0 or 1
[0028] Based on the third problem model, the Monte Carlo algorithm is used to solve the polynomial coefficient θ.
[0029] Furthermore, after the step of identifying the independent loads included in the user's comprehensive electricity consumption data, the step of updating the user's load feature library is also included, which is specifically as follows:
[0030] Determine whether the sparse representation of the current overcomplete dictionary for the newly added single load feature meets the error threshold; update the current overcomplete dictionary only when the sparse representation error is greater than the error threshold;
[0031] When updating the current over-complete dictionary, the newly added single load feature is used as a valid sample, and the values of each element in the sparse matrix of the newly added single load under the current over-complete dictionary are observed. Only when the element is not zero, the corresponding atom of the over-complete dictionary is updated using K-SVD, and the iterative calculation is performed until the maximum number of iterations is reached or the error threshold is met; if the error threshold cannot be met by updating the dictionary with the newly added single load feature as a valid sample, all load features are used to update the over-complete dictionary.
[0032] Furthermore, after the step of identifying the independent loads included in the user's comprehensive electricity consumption data, the step of updating the user's load feature library is also included, which is specifically as follows:
[0033] Input: Current user load characteristic database Current user load characteristic database Sparse dictionary of The sparse representation X of the user load feature library L under the dictionary D, and the new load feature l M+1 , the error threshold ε, and the maximum number of iterations J;
[0034] Initialization: D 0 = D, L = [L, l M+1 ,
[0035] Step 1: Solve using the orthogonal matching pursuit algorithm That is, the sparse representation matrix of the new load feature l under the sparse dictionary D M+1
[0036] Step 2: Calculate If then the dictionary D does not need to be updated, go to Step 7; if execute Step 3;
[0037] Step 3: Update the dictionary D
[0038] a) Initialization: i = 1;
[0039] b) If go to Step c; if update the i-th atom d of the dictionary D j and X i ; j
[0040] c) If i < P, i = i + 1, return to Step b; if i ≥ P, let D j+1 = D j , output the dictionary D j+1 ;
[0041] Step 4: Solve L = D j+1 X j+1 , output X j+1 ;
[0042] Step 5: If ||L - D j+1 X j+1 || > ε and j + 1 ≤ J, j = j + 1, return to Step 3; if ||L - D j+1 X j+1 || ≤ ε and j + 1 ≤ J, output D j+1 , complete the fast online dictionary update, go to Step 7; if ||L - D j+1 X j+1 ‖ > ε and j + 1 > J, go to Step 6;
[0043] Step 6: Update all atoms of the dictionary D using K - SVD and X j+1 j+1 , until the maximum number of iterations 2J is reached or the error threshold ε is met;
[0044] Step 7: End.
[0045] A second aspect of the present invention provides a user load identification fusion terminal, comprising:
[0046] A communication module, used for communicating with multiple types of sensors connected to a low-voltage distribution network by plugging;
[0047] The core processing module is used to obtain data from multiple types of sensors of the low-voltage distribution network through the communication module, and to identify user loads based on the data from the multiple types of sensors using the user load identification method as described above; wherein the data from the multiple types of sensors are comprehensive electricity consumption data of users.
[0048] According to a third aspect of the present invention, there is provided a user load identification device, comprising:
[0049] A feature library acquisition module is used to acquire a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding;
[0050] A problem conversion module is used to obtain comprehensive power consumption data of users; a load identification method based on integer programming is adopted, based on the user load feature library, to convert the decomposition problem of comprehensive power consumption data of users into an integer programming problem;
[0051] The solution module is used to solve the integer programming problem and identify the independent loads included in the user's comprehensive electricity consumption data.
[0052] According to a fourth aspect of the present invention, an electronic device is provided, comprising a processor and a memory, wherein the processor is configured to execute a computer program stored in the memory to implement the user load identification method as described above.
[0053] According to a fifth aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the user load identification method as described above is implemented.
[0054] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0055] The present invention provides a user load identification method based on sparse coding and integer programming. First, a user load feature library is constructed using sparse coding technology, and the load characteristics of electrical equipment are stored comprehensively and compactly through the representation of an over-complete dictionary and a sparse matrix. Then, in the load identification process, a load identification method based on integer programming is used to convert the decomposition problem of the user's comprehensive electricity consumption data into an integer programming problem. By solving this problem, the independent loads contained in the user's comprehensive electricity consumption data are identified, and the multivariate load feature information is further integrated, the information redundancy is reduced, the algorithm complexity is simplified, and the efficiency and accuracy of user load identification are improved. A user load identification device, an electronic device, and a computer-readable storage medium provided by the present invention also solve the problems raised in the background technology section.
[0056] In addition, the present invention also proposes a method for updating a feature library, which can quickly update and expand the user load feature library when new electrical equipment is added, thereby ensuring the accuracy and robustness of the recognition algorithm. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] The drawings constituting a part of the present 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:
[0058] Figure 1 A schematic diagram of a flow chart of a method for identifying user load according to an embodiment of the present invention;
[0059] Figure 2 This is a simplified structural diagram of a user load identification fusion terminal according to an embodiment of the present invention;
[0060] Figure 3 This is a structural block diagram of a user load identification device according to an embodiment of the present invention;
[0061] Figure 4 The present invention is a structural block diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0062] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.
[0063] The following detailed description is an exemplary description, which is intended to provide further detailed description of the present invention. Unless otherwise specified, all technical terms used in the present invention have the same meaning as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present invention are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention.
[0064] Example 1
[0065] like Figure 1 As shown, a user load identification method is characterized by comprising the following steps:
[0066] S1. Obtain a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding;
[0067] S2. Obtaining comprehensive electricity consumption data of users; using a load identification method based on integer programming, based on the user load feature library, converting the decomposition problem of comprehensive electricity consumption data of users into an integer programming problem;
[0068] S3. Solve the integer programming problem and identify the independent loads included in the user's comprehensive electricity consumption data.
[0069] The user load identification method provided by this solution can more accurately decompose the user's comprehensive electricity consumption data and effectively identify the independent loads by constructing a user load feature library based on sparse coding and using integer programming methods for load identification. In addition, the algorithm complexity is simplified, while the efficiency and accuracy of user load identification are improved.
[0070] To further explain the present solution, in a specific embodiment, a method for identifying user load is also provided, comprising the following steps:
[0071] S100: Acquire a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding.
[0072] In order to integrate multi-element load feature information, reduce information redundancy, and simplify algorithm complexity, this solution uses a sparse representation algorithm to characterize the user load feature library. Among them, each load feature in the user load feature library can be obtained by relying on a public data set, or the user load can be registered through a preliminary experiment, which will not be described in detail in this solution. In addition, each load feature in the user load feature library can be a combination of one or more features, such as current waveform features, harmonic features, active power, reactive power, etc., which is not restricted in this solution.
[0073] Specifically, the user load feature library is characterized based on sparse coding, including:
[0074] Assume that a user load characteristic database is: in, is the feature of the i-th load in the user load feature library, the feature dimension is N×1, and there are M types of loads in the user load feature library;
[0075] Assume that the over-complete dictionary used to characterize the user load feature library is Contains P groups of basis vectors, and P>>N;
[0076] The user load feature library based on sparse coding is expressed as:
[0077] L≈DX (1)
[0078] in, It is the sparse matrix of user load feature library L.
[0079] This scheme uses the greedy algorithm principle to iteratively solve the variables D and X: fix D and use the orthogonal matching pursuit algorithm to solve X; fix X and use the K-SVD algorithm to solve D. Then the user load feature library L can be uniformly represented by an overcomplete dictionary and a sparse matrix.
[0080] S200, obtaining comprehensive electricity consumption data of users; using a load identification method based on integer programming, based on the user load feature library, converting the decomposition problem of the comprehensive electricity consumption data of users into an integer programming problem.
[0081] The applicant needs to explain that the purpose of load identification is to decompose the user's comprehensive electricity consumption data, combine it with the user's load feature library built in advance, and identify the independent loads contained in the user's comprehensive electricity consumption data.
[0082] In this scheme, the user's comprehensive electricity consumption data is a linear combination of multiple loads. Therefore, in order to further integrate the characteristic information of multiple loads, reduce information redundancy, and simplify the complexity of the algorithm, this scheme designs a load identification method based on integer programming, converts the decomposition problem into an integer programming problem, and uses a sparse representation algorithm to characterize the user load feature library L.
[0083] Specifically, a load identification method based on integer programming is adopted, and based on the user load feature library, the decomposition problem of the user's comprehensive power consumption data is converted into an integer programming problem, including:
[0084] Construct the first problem model:
[0085]
[0086] in, is the sparse representation of the user load feature library L under the dictionary D; θ=[θ1,θ2,...θ M ] T is the polynomial coefficient, which is used to characterize each load operation state; X Y is the sparse representation of Y under dictionary D, Y = DX Y , The comprehensive load characteristics of the user's comprehensive electricity consumption data are compared with the individual load characteristics in the user load characteristic library. i have the same characteristics;
[0087] The first problem model is expressed as the second problem model:
[0088]
[0089] Among them, in the practical problem of load identification, θ i =0 means that the i-th load is not a component of the current user's comprehensive power consumption data, that is, the i-th load is not in operation; otherwise, θ i =1 indicates that the i-th load is a component of the current user's comprehensive electricity consumption data, that is, the i-th load is in operation.
[0090] S300, solving the integer programming problem, and identifying independent loads included in the user's comprehensive electricity consumption data.
[0091] Specifically, solving the integer programming problem and identifying the independent loads contained in the user's comprehensive electricity consumption data includes:
[0092] Let Q = X T X, C = -2X Y T X; due to θ i =0 / 1, then we know that θ i =θ i 2 ;
[0093] Let Q = Q + diag (C), then the second problem model is further expressed as the third problem model:
[0094]
[0095] At this point, this scheme converts the load identification problem into a nonlinear integer programming problem. Based on the third problem model, the Monte Carlo algorithm is used to solve the polynomial coefficient θ.
[0096] In a preferred embodiment, after the step of identifying the independent loads contained in the user's comprehensive electricity consumption data, in actual applications, the user load feature library needs to increase the load types according to the user's actual situation. When the load types increase, the sparse representation dictionary of the user load feature library also needs to be updated accordingly. In order to improve the efficiency of dictionary updates and reduce the complexity of the dictionary update algorithm, this solution proposes a fast online dictionary update method based on K-SVD, thereby meeting the requirements of the edge acquisition terminal for a simple, easy to deploy, and easy to update algorithm for the load identification function.
[0097] The steps for updating the user load feature database are as follows:
[0098] Determine whether the sparse representation of the current overcomplete dictionary for the newly added single load feature meets the error threshold; update the current overcomplete dictionary only when the sparse representation error is greater than the error threshold;
[0099] When updating the current overcomplete dictionary, using the newly added single load feature as a valid sample, observe the values of each element in the sparse matrix of the newly added single load under the current overcomplete dictionary. Only when the element is not zero, use K-SVD to update the corresponding atom of the overcomplete dictionary, and perform iterative calculations until the maximum number of iterations is reached or the error threshold is satisfied; if updating the dictionary using the newly added single load feature as a valid sample cannot meet the error threshold, then update the overcomplete dictionary using all load features.
[0100] In a preferred embodiment, after the step of identifying the independent loads included in the user's comprehensive power consumption data, it further includes the step of updating the user load feature library, specifically as follows:
[0101] Input: The current user load feature library The current user load feature library The sparse dictionary of The sparse representation X of the user load feature library L under the dictionary D, the newly added load feature l M+1 , the error threshold ε, the maximum number of iterations J;
[0102] Initialization: D 0 = D, L = [L, l M+1 , j = 0;
[0103] Step 1: Use the orthogonal matching pursuit algorithm to solve That is, the sparse representation matrix of the newly added load feature l under the sparse dictionary D M+1 of
[0104] Step 2: Calculate If then the dictionary D does not need to be updated, go to Step 7; if execute Step 3;
[0105] Step 3: Update the dictionary D
[0106] a) Initialization: i = 1;
[0107] b) If go to Step c; if use the K-SVD method to update the i-th atom d of the dictionary D j and X i ; j ;
[0108] c) If i < P, i = i + 1, return to Step b; if i ≥ P, let D j+1 = D j , output the dictionary D j+1 ;
[0109] Step 4: Solve L = D using the orthogonal matching pursuit algorithm j+1 X j+1 , output X j+1 ;
[0110] Step 5: If ‖LD j+1 X j+1 ‖>ε and j+1≤J, j=j+1, return to step 3; if ‖LD j+1 X j+1 ‖≤ε and j+1≤J, output D j+1 , complete the fast online dictionary update and go to step 7; if ‖LD j+1 X j+1 ‖>ε and j+1>J, go to step 6;
[0111] Step 6: Update dictionary D using K-SVD j+1 All atoms and X j+1 , until the maximum number of iterations 2J is reached or the error threshold ε is met;
[0112] Step 7: End.
[0113] In an optional embodiment, the user load identification method may include:
[0114] Step 100: construct an overcomplete dictionary and sparse matrix of the user load feature library;
[0115] Step 200: Decomposing the user's comprehensive electricity consumption data for load identification
[0116] Input to obtain the user's current comprehensive electricity consumption data;
[0117] Using the overcomplete dictionary of the user load feature library, the sparse matrix, and the user's comprehensive electricity consumption data, the Monte Carlo algorithm is used to solve formula (4);
[0118] Output the operating status of each load.
[0119] Step 300: When a user adds a new load type, a fast online dictionary update method can be used to update the sparse dictionary of the existing user load feature library to facilitate subsequent accurate load identification.
[0120] A user load identification method proposed in an embodiment of the present invention adopts non-invasive technology. First, a sparse representation algorithm is used to characterize the user load feature library, and an over-complete dictionary and sparse matrix of the user load feature library are constructed. Secondly, the load identification problem of the user's comprehensive electricity consumption data is converted into an integer programming problem based on sparse representation. By solving the integer programming problem, the decomposition of the user's comprehensive electricity consumption data is realized and the load identification effect is achieved. Finally, a fast online dictionary update method is proposed to deal with the online update problem of the user load feature library. The non-invasive load identification method proposed in this scheme has the characteristics of simple algorithm, easy deployment, and easy update.
[0121] Example 2
[0122] Embodiment 2 of the present invention is based on Embodiment 1 and provides a user load identification fusion terminal, including:
[0123] A communication module, used for communicating with multiple types of sensors connected to a low-voltage distribution network by plugging;
[0124] The core processing module is used to obtain data from multiple types of sensors of the low-voltage distribution network through the communication module, and to identify user loads based on the data from the multiple types of sensors using the user load identification method as described above; wherein the data from the multiple types of sensors are comprehensive electricity consumption data of users.
[0125] Further, such as Figure 2 As shown, a user load identification fusion terminal provided in Example 2 of the present invention specifically includes a display module, a communication module, a core processing module, an exchange board, a power board, etc. Among them, the core processing module is integrated on the main control board, and is mainly used to identify the user's load based on the user load identification method of Example 1. The display module is used for collected data, load identification results, terminal status, etc. The exchange board is mainly used to sample the user's comprehensive load data.
[0126] In an optional embodiment, the user load identification fusion terminal provided in Embodiment 2 of the present invention is a standardized, universal, miniaturized fusion terminal, which is deployed with the user load identification method of Embodiment 1. The fusion terminal has the characteristics of small installation size, high reliability, low power consumption, and rich interfaces, and can realize hardware platformization and software APPization, support the interconnection and intercommunication of multiple types of sensors in the low-voltage distribution network and plug-and-play of communication modules, and realize rapid deployment and installation of acquisition terminals, standardized data access, and localized intelligent analysis of edge computing.
[0127] Example 3
[0128] like Figure 3 As shown, based on the same inventive concept as the above embodiment, the present invention also provides a user load identification device, comprising:
[0129] A feature library acquisition module is used to acquire a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding;
[0130] A problem conversion module is used to obtain comprehensive power consumption data of users; a load identification method based on integer programming is adopted, based on the user load feature library, to convert the decomposition problem of comprehensive power consumption data of users into an integer programming problem;
[0131] The solution module is used to solve the integer programming problem and identify the independent loads included in the user's comprehensive electricity consumption data.
[0132] Example 4
[0133] like Figure 4 As shown, the present invention also provides an electronic device 100 for implementing the user load identification method of embodiment 1;
[0134] The electronic device 100 includes a memory 101 , at least one processor 102 , a computer program 103 stored in the memory 101 and executable on the at least one processor 102 , and at least one communication bus 104 .
[0135] The memory 101 may be used to store a computer program 103 . The processor 102 implements the steps of a user load identification method in Example 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101 .
[0136] The memory 101 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data (such as audio data) created according to the use of the electronic device 100, etc. In addition, the memory 101 may include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
[0137] At least one processor 102 may be a central processing unit (CPU), or 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 processor 102 may be a microprocessor or any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and uses various interfaces and lines to connect various parts of the entire electronic device 100.
[0138] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a user load identification method, and the processor 102 may execute the plurality of instructions to implement:
[0139] Acquire a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding;
[0140] Obtaining comprehensive electricity consumption data of users; using a load identification method based on integer programming, based on the user load feature library, converting the decomposition problem of the comprehensive electricity consumption data of users into an integer programming problem;
[0141] Solve the integer programming problem and identify the independent loads included in the user's comprehensive electricity consumption data.
[0142] Example 5
[0143] If the module / unit integrated in the electronic device 100 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned method embodiments when executed by the processor. Among them, 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. Computer-readable media may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory and read-only memory (ROM, Read-Only Memory).
[0144] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take 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.) containing computer-usable program code.
[0145] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks 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 produce 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 1 A device that provides the functions specified in a block or multiple blocks.
[0146] 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.
[0147] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0148] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.
Claims
1. A method for identifying user load, characterized in that: The steps include: Acquire a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding; Obtaining comprehensive electricity consumption data of users; using a load identification method based on integer programming, based on the user load feature library, converting the decomposition problem of the comprehensive electricity consumption data of users into an integer programming problem; Solve the integer programming problem and identify the independent loads included in the user's comprehensive electricity consumption data.
2. The user load identification method according to claim 1, characterized in that: The user load feature library is characterized based on sparse coding, and includes: Assume that a user load characteristic library is: in, is the feature of the i-th load in the user load feature library, the feature dimension is N×1, and there are M types of loads in the user load feature library; Assume that the over-complete dictionary used to characterize the user load feature library is Contains P groups of basis vectors, and P>>N; The user load feature library based on sparse coding is expressed as: L≈DX; where, It is the sparse matrix of user load feature library L.
3. The user load identification method according to claim 2, characterized in that: A load identification method based on integer programming is adopted, and based on the user load feature library, the decomposition problem of the user's comprehensive power consumption data is converted into an integer programming problem, including: Construct the first problem model: stθ i =0 or 1 in, is the sparse representation of the user load feature library L under the dictionary D; θ=[θ1,θ2,...θ M ] T is the polynomial coefficient, which is used to characterize each load operation state; X Y is the sparse representation of Y under dictionary D, Y = DX Y , The comprehensive load characteristics of the user's comprehensive electricity consumption data are compared with the individual load characteristics in the user load characteristic library. i have the same characteristics; The first problem model is expressed as the second problem model: minθ T X T Xθ-2X Y T Xθ+X Y T X Y STθ i =0 or 1 Among them, θ i =0 means that the i-th load is not a component of the current user's comprehensive electricity consumption data; θ i =1 indicates that the i-th load is a component of the current user's comprehensive electricity consumption data.
4. The user load identification method according to claim 3, characterized in that: Solve the integer programming problem and identify the independent loads contained in the user's comprehensive electricity consumption data, including: Let Q = X T X, C = -2X Y T X; due to θ i =0 / 1, then we know that θ i =θ i 2 ; Let Q = Q + diag (C), then the second problem model is further expressed as the third problem model: minθ T Qθ stθ i =0 or 1 Based on the third problem model, the Monte Carlo algorithm is used to solve the polynomial coefficient θ.
5. The user load identification method according to claim 2, characterized in that: After the step of identifying the independent loads included in the user's comprehensive electricity consumption data, the step of updating the user load feature library is also included, which is specifically as follows: Determine whether the sparse representation of the current overcomplete dictionary for the newly added single load feature meets the error threshold; update the current overcomplete dictionary only when the sparse representation error is greater than the error threshold; When updating the current over-complete dictionary, the newly added single load feature is used as a valid sample, and the values of each element in the sparse matrix of the newly added single load under the current over-complete dictionary are observed. Only when the element is not zero, the corresponding atom of the over-complete dictionary is updated using K-SVD, and the iterative calculation is performed until the maximum number of iterations is reached or the error threshold is met; if the error threshold cannot be met by updating the dictionary with the newly added single load feature as a valid sample, all load features are used to update the over-complete dictionary.
6. The user load identification method according to claim 2, characterized in that: After the step of identifying the independent loads included in the user's comprehensive electricity consumption data, the step of updating the user load feature library is also included, which is specifically as follows: Input: Current user load characteristic database Current user load characteristic database Sparse dictionary of The sparse representation X of the user load feature library L under the dictionary D, the newly added load feature l M+1 , error threshold ε, maximum number of iterations J; Initialization: D 0 =D,L=[L,l M+1 ], j = 0; Step 1: Solve using the orthogonal matching pursuit algorithm That is, the new load feature l under the sparse dictionary D M+1 The sparse representation matrix Step 2: Calculation like Then dictionary D does not need to be updated, go to step 7; if Execute step 3; Step 3: Update dictionary D a) Initialization: i = 1; b) If Go to step c; if Update dictionary D using K-SVD method j The i-th atom d i and X j ; c) If i < P, set i = i + 1 and return to step b; if i ≥ P, set D j+1 = D j , and output the dictionary D j+1 ; Step 4: Solve L = D using the orthogonal matching pursuit algorithm j+1 X j+1 , output X j+1 ; Step 5: If ‖LD j+1 X j+1 ‖>ε and j+1≤J, j=j+1, return to step 3; if ‖LD j+1 X j+1 ‖≤ε and j+1≤J, output D j+1 , complete the fast online dictionary update and go to step 7; if || LD j+1 X j+1 ||>ε and j+1>J, go to step 6; Step 6: Update dictionary D using K-SVD j+1 All atoms and X j+1 , until the maximum number of iterations 2J is reached or the error threshold ε is met; Step 7: End.
7. A user load identification fusion terminal, characterized in that: include: A communication module, used for communicating with multiple types of sensors connected to a low-voltage distribution network by plugging; The core processing module is used to obtain data from multiple types of sensors of the low-voltage distribution network through the communication module, and to perform user load identification based on the data from the multiple types of sensors using the user load identification method as described in any one of claims 1 to 6; wherein the data from the multiple types of sensors are comprehensive electricity consumption data of users.
8. A user load identification device, characterized in that: include: A feature library acquisition module is used to acquire a pre-built user load feature library; wherein the user load feature library is characterized based on sparse coding; A problem conversion module is used to obtain comprehensive power consumption data of users; a load identification method based on integer programming is adopted, based on the user load feature library, to convert the decomposition problem of comprehensive power consumption data of users into an integer programming problem; The solution module is used to solve the integer programming problem and identify the independent loads included in the user's comprehensive electricity consumption data.
9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the user load identification method as described in any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the user load identification method according to any one of claims 1 to 6 is implemented.
