A power distribution network harmonic state estimation method for lack of measurement and parameter ambiguity, electronic equipment and computer readable storage medium

Abstract

The application discloses a kind of power distribution network's harmonic state estimation method, electronic equipment and computer readable storage medium for lack of measurement and parameter ambiguity, belong to power distribution network power quality and state estimation technical field.The application is aimed at the problem that multiple solutions exist in harmonic state estimation and convergence is poor due to insufficient measurement configuration and line parameter ambiguity of power distribution network, constructs space-frequency joint estimation model based on compressed sensing theory using space-frequency dual sparsity of harmonic source;Error coupling and amplification caused by parameter deviation are eliminated by introducing spatial whitening mapping;Dimensionality reduction is solved by extracting prior support set, and dynamic updating of support set is realized by combining residual compressed sensing and double threshold detection.The application significantly reduces the requirement for measurement configuration and parameter accuracy, and can realize high-precision and high-robust harmonic state estimation in the scene of measurement deficiency and parameter ambiguity, and is suitable for power distribution network harmonic evaluation and harmonic source positioning.

Description

Technical Field

[0001] This invention relates to the field of power quality analysis and state estimation technology for distribution networks, and particularly to a method, electronic device, and computer-readable storage medium for harmonic state estimation of distribution networks with limited vector measurements and ambiguous parameters. Background Technology

[0002] With the large-scale integration of nonlinear loads such as high-proportion renewable energy and power electronic equipment into the distribution network, voltage and current waveforms of the power grid are distorted, and harmonic pollution problems are becoming increasingly prominent. Harmonic state estimation is a core technical means to assess the harmonic level of the power grid, locate harmonic sources, and support power quality management. It calculates the harmonic operating status of each node in the entire network by combining limited measurement data with network topology and parameter models.

[0003] Existing methods for estimating the harmonic state of distribution networks are generally based on the idealized assumptions of "observable across the entire network" and "precisely known line parameters": on the one hand, the system is required to be equipped with sufficient measuring devices so that the measurement equations are positive definite or overdetermined equations to obtain a unique solution; on the other hand, network parameters such as line impedance are required to be completely consistent with the actual operating conditions to ensure the numerical stability of the measurement matrix.

[0004] However, in actual engineering scenarios, both of the above assumptions are difficult to satisfy. First, the number of nodes in a distribution network is enormous, and due to economic constraints, it is impossible to deploy synchronous phasor measurement units at all nodes. The configuration of measurement devices is severely insufficient, resulting in underdetermined equations with infinitely many solutions, making it impossible for traditional algorithms to obtain accurate estimation results. Second, the branch lines of a distribution network are complex and usually do not have distance protection devices. In daily operation and maintenance, there is a lack of on-site verification of line parameters. Theoretical parameters can only be calculated based on the line type and design length. However, the actual line parameters are affected by factors such as aging, temperature drift, and skin effect, resulting in significant deviations. This leads to singular values ​​in the measurement matrix, causing poor convergence of the algorithm or even complete failure to converge, ultimately resulting in the failure of harmonic state estimation.

[0005] To address the problem of insufficient measurement, some studies have introduced compressed sensing theory into harmonic state estimation, using the sparsity of harmonic sources to solve underdetermined equations. However, most existing methods do not simultaneously consider the model errors caused by line parameter ambiguity, and lack an effective dynamic support set update mechanism. In scenarios with parameter deviations, problems such as error amplification and pseudo-harmonic source interference are prone to occur, and the estimation accuracy and robustness still cannot meet the needs of practical engineering. Summary of the Invention

[0006] To address the technical problems of existing technologies for harmonic state estimation in distribution networks, which rely on observable conditions of the entire network and precise line parameters, resulting in multiple solutions, poor convergence, and low estimation accuracy in real-world scenarios with scarce measurements and ambiguous parameters, this invention provides a method, electronic device, and computer-readable storage medium for harmonic state estimation in distribution networks with scarce measurements and ambiguous parameters.

[0007] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: The present invention utilizes the natural sparsity of the spatial location and harmonic frequency of harmonic sources in the distribution network, and constructs a space-frequency joint estimation framework based on compressed sensing theory, overcoming the limitation of observable conditions across the entire network; it introduces spatial whitening mapping to handle the uncertainty of line impedance parameters, suppressing error amplification caused by parameter deviations; it reduces the solution dimension by extracting prior support sets, and achieves dynamic updating of the support sets by combining residual compressed sensing and dual-threshold closed-loop detection, thus achieving high-precision and highly robust distribution network harmonic state estimation under the dual unfavorable conditions of scarce measurements and ambiguous parameters.

[0008] A method for harmonic state estimation in distribution networks with limited surface vector measurement and ambiguous parameters includes the following steps:

[0009] S1 Model Establishment: Based on the total number of nodes in the distribution network and the harmonic frequencies to be analyzed, a global state vector containing spatial and frequency dimensions is established. The single-frequency sensing sub-matrix is ​​calculated based on Kirchhoff's laws. The block diagonal structure space-frequency joint sensing matrix is ​​constructed through Kronecker product operation. A joint measurement equation considering the deviation of line impedance parameters is established.

[0010] S2 Key Prior Parameter Acquisition and Spatial Whitening Mapping: Extract the physical parameters of the distribution network lines, calculate the theoretical network matrix corresponding to each harmonic after skin effect correction, construct the comprehensive error covariance matrix in combination with the noise parameters of the measurement equipment, perform eigenvalue decomposition on the comprehensive error covariance matrix to obtain the whitening matrix, and use the whitening matrix to perform linear transformation on the joint measurement equation to obtain the whitened perception matrix and measurement vector.

[0011] S3 Support Set Extraction: Based on the harmonic state estimation results and the corresponding estimation error covariance matrix of the previous measurement time, calculate the adaptive detection threshold of each potential harmonic source node, filter the node indexes whose estimated amplitudes at the previous time meet the threshold requirements, and form the prior support set for the current time.

[0012] S4 Residual Signal Stripping and Burst Source Detection: Based on the prior support set, the corresponding column vectors are extracted from the whitened perception matrix to form an active submatrix. The state estimates of the harmonic sources in the prior support set are solved by unbiased least squares operation. The measurement responses corresponding to the state estimates are stripped from the whitened measurement vectors to obtain the residual signal.

[0013] S5 Space-Frequency Joint Residual Compressed Sensing Optimization for New Burst Sources: Energy threshold detection is performed on the residual signal. If it is determined that there is a new harmonic source, space-frequency joint residual compressed sensing optimization is performed in the index space outside the prior support set using hybrid norm constraints to solve for the preliminary estimated state of the new harmonic source.

[0014] S6 High Robust Dual Threshold Detection and Support Set Update: A new amplitude threshold detection is performed on the preliminary estimated state of newly added harmonic sources, and qualified nodes are added to the temporary support set; based on the temporary support set, global unbiased least squares estimation is re-executed, and failed harmonic sources are removed by amplitude drop detection, and the final support set is updated to obtain the final support set, and the network-wide harmonic state estimation result at the current moment is output.

[0015] Furthermore, in step S1, the global state vector is reduced to a one-dimensional vector by combining the injected current vectors of all network nodes under each harmonic frequency; the joint measurement equation combines the line parameter perturbation and the measurement equipment noise into an equivalent comprehensive correlation noise, which is used as the error term of the measurement equation.

[0016] Further, in step S2, the theoretical network matrix includes the node admittance matrix and current transfer matrix corresponding to each harmonic; the comprehensive error covariance matrix is ​​constructed by combining the prior parameter error variance and the state quantity at the previous moment, and the relative error standard deviation of the prior parameter error variance is 5%.

[0017] Furthermore, in step S2, the spatial whitening mapping is used to eliminate the correlation of measurement channels caused by the topological coupling of the distribution network, balance the dimensional differences between voltage measurement and current measurement, and convert the equivalent integrated correlated noise into spatially uncorrelated standard white noise, thereby suppressing the multi-level amplification of errors caused by parameter deviation.

[0018] Furthermore, step S3 specifically includes:

[0019] S3.1 Based on the whitened perceptron submatrix and the corresponding support set of the previous time step, the covariance matrix of the harmonic state estimation error of the previous time step is calculated analytically.

[0020] S3.2 For each potential harmonic source node, extract the standard deviation of its state estimation error, and combine it with the statistical confidence multiplier to generate an adaptive detection threshold specific to that node.

[0021] S3.3 compares the estimated harmonic amplitude of each node at the previous time step with the corresponding adaptive detection threshold, and includes the node index with amplitude greater than the threshold in the prior support set at the current time step.

[0022] Furthermore, in step S4, the unbiased least squares operation is solved using Moore-Penrose pseudo-inverse to correct the magnitude shrinkage to zero bias caused by pure compressed sensing.

[0023] Furthermore, in step S5, the hybrid norm constraint includes Norm and norm, where Norm constraints on the overall sparsity of nodes across the entire frequency band Norm constraints are applied to the local sparsity of non-zero nodes in the frequency dimension; the tolerance boundary of the optimization problem is determined based on the chi-square distribution.

[0024] Furthermore, in step S6, the new amplification threshold is a multiple of the worst-case reconstruction error of compressed sensing, and the coefficient ranges from 3 to 5, corresponding to the confidence level requirement under normal distribution.

[0025] Further, in step S6, the amplitude drop detection process is as follows: after completing the global unbiased least squares estimation based on the temporary support set, if the estimated amplitude of the original harmonic source drops below its corresponding adaptive detection threshold, the node is removed from the support set, and the unbiased least squares estimation and amplitude drop detection are repeated until no node is removed.

[0026] Furthermore, the final support set obtained in step S6 will be passed to the next measurement time section as the basis for calculating the a priori support set at the next moment, forming a dynamic continuous tracking closed loop of the distribution network harmonic state.

[0027] An electronic device, characterized in that the device comprises:

[0028] At least one processor; and

[0029] A memory communicatively connected to the at least one processor; wherein,

[0030] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the operation as described in claim 1.

[0031] A computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions, the computer instructions being used to cause a processor to execute and implement the claims of claim 1.

[0032] Compared with the prior art, the present invention has the following significant advantages:

[0033] 1) Overcoming the limitations of insufficient measurement and solving the problem of multiple solutions to underdetermined equations.

[0034] This invention fully utilizes the natural sparsity of both the spatial distribution of harmonic sources and the harmonic frequency in the distribution network. Based on compressed sensing theory, it constructs a joint space-frequency estimation framework, which can solve the under-quantitative measurement equations without satisfying the observability condition of the entire network. At the same time, by extracting the prior support set of adjacent time moments, it transforms high-dimensional blind optimization into low-dimensional accurate solution, which greatly reduces the rigid requirement for the number of measurement channels and effectively solves the problems of multiple solutions and non-convergence caused by insufficient measurement.

[0035] 2) Adapt to scenarios with blurred parameters and suppress multi-level error amplification.

[0036] This invention incorporates line parameter deviations into the measurement equation modeling, decouples the covariance matrix of the comprehensive error through spatial whitening mapping, eliminates the correlation of measurement channels caused by strong topological coupling of the distribution network, balances the dimensional differences of different measurement types, and transforms parameter perturbation and measurement noise into spatially uncorrelated standard white noise, avoiding error superposition and amplification caused by parameter ambiguity, and improving the estimation stability under uncertain model scenarios.

[0037] 3) Dynamic closed-loop tracking improves estimation robustness.

[0038] This invention designs a support set dynamic update mechanism based on residual compressed sensing and dual threshold detection: new harmonic sources are identified by residual energy detection, and invalid harmonic sources are eliminated by amplitude drop detection, thus realizing real-time dynamic tracking of harmonic sources; at the same time, a solution strategy combining unbiased least squares and compressed sensing is adopted to correct the amplitude shrinkage bias of the pure compressed sensing algorithm, which has both sparse solution capability and high-precision estimation characteristics, effectively avoiding overfitting and pseudo-harmonic source interference, and the estimation results are more robust;

[0039] 4) The algorithm is highly efficient and has strong engineering applicability.

[0040] This invention relies on a priori support set to significantly compress the solution space. It triggers high-dimensional optimization operations only when a new harmonic source is detected, and only low-dimensional least squares calculations are required during routine tracking, thus significantly improving the computational efficiency of the algorithm. At the same time, the threshold selection is based on rigorous derivation of statistical distribution, avoiding blind manual parameter tuning and adapting to the engineering application needs of actual power distribution networks. Attached Figure Description

[0041] Figure 1 This is a flowchart of the method of the present invention;

[0042] Figure 2 This is a schematic diagram of the electronic device used in the method for harmonic state estimation of distribution networks with limited surface vector measurement and ambiguous parameters according to the present invention.

[0043] Figure 2Chinese reference numerals: 1. Electronic device; 11. Processor; 12. ROM; 13. RAM; 14. Bus; 15. Input / output (I / O) interface; 16. Input unit; 17. Output unit; 18. Storage unit; 19. Communication unit. Detailed Implementation

[0044] To enable those skilled in the art to better understand the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention. It should be understood that the various steps described in the method embodiments of the present invention can be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0045] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0046] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, any variations of the terms "comprising" and "having," etc., are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0047] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0048] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0049] Example 1

[0050] like Figure 1 As shown in the figure, the harmonic state estimation method for distribution networks with scarce surface vector measurements and ambiguous parameters provided in this embodiment has the following specific implementation steps:

[0051] Step S1: Model Building

[0052] Assume the distribution network system has a total of Each node, pay attention to One potential harmonic frequency, in At any time, the entire network includes One voltage measurement channel and One branch current measurement channel, total measurement dimensions Due to the scarcity of measuring devices, .

[0053] S1.1 Establish the global state vector to be determined

[0054] Based on the nodes of interest and harmonic frequencies, establish the state matrix to be determined. , This represents a joint harmonic injection matrix that incorporates both spatial and frequency attributes, and it has a total of Columns, each column contains the entire network The injected current at a specific frequency at each node.

[0055] Let its column vector be represented as Each of them The mathematical expression for a matrix is:

[0056]

[0057] Dimensionality reduction is performed to reduce the spatial dimensions. and frequency dimension Jointly reduce the dimensionality to a one-dimensional global state vector. :

[0058]

[0059] S1.2: Establishing a joint space-frequency compressed sensing matrix

[0060] Based on Kirchhoff's laws, solve for the measurement matrix of a single frequency. .

[0061] The Kronecker product is used to combine the independent frequency sensing sub-matrices into a global block diagonal joint sensing matrix. The combination process is as follows:

[0062] set up For the first A standard basis vector with elements 1 and the rest 0, then It is The matrix, only in the first... The position is 1, and all other positions are 0;

[0063] Joint sensing matrix It can be strictly constructed using the summation formula of the Kronecker product:

[0064]

[0065] According to the rules of the Kronecker product, expanding it yields a strictly block diagonal matrix form:

[0066]

[0067] S1.3: Establish the joint measurement equation considering the uncertainty of the impedance model

[0068] Considering the inaccuracies of the simulated impedance model, a rigorous joint measurement equation is established as follows:

[0069]

[0070] in, This is a combined measurement vector of multi-frequency voltage and current. The parameter inaccuracy matrix perturbation caused by line aging and temperature drift; The inherent broadband noise of the current transformer and PMU; the two are combined into an equivalent synthesized correlated noise. .

[0071] S2 Key Prior Parameter Acquisition and Spatial Whitening Mapping

[0072] The prerequisite for implementing residual compressed sensing is to correctly acquire and quantitatively use various physical prior error parameters to construct a whitening operator, thereby eliminating the strong correlation caused by inaccurate network parameters and dimensional differences between multiple source sensors.

[0073] S2.1 Obtaining Prior Parameters

[0074] Physical information such as line material and length is extracted from the power distribution management system and GIS database. The fundamental impedance is calculated and corrected for the skin effect. The theoretical nodal admittance / impedance matrix and current transfer matrix for each harmonic are generated, and the matrix is ​​constructed. .

[0075] By consulting the datasheets for the accuracy levels of the current transformers and PMUs and statically assigning values, a matrix was constructed. .

[0076] The variance of the prior parameter error is an empirical value, which can conservatively represent the standard deviation of the relative error of the impedance matrix. Set it to 5%.

[0077] S2.2 Covariance Matrix Construction and Whitening Processing

[0078] Using the prior variance and the state at the previous time step Analyze and construct the covariance matrix of the comprehensive error. :

[0079]

[0080] right Perform eigenvalue decomposition to obtain the whitening matrix. .Will Multiplying the measurement equation by left-hand side yields the whitening mapping system:

[0081]

[0082] in ,

[0083] S2.2 not only balances the dimensional differences between voltage and current, eliminating the measurement channel correlation caused by strong topological coupling in the distribution network, but also forcibly transforms the complexly distributed comprehensive error into spatially uncorrelated standard white noise. This avoids the multi-level amplification of errors caused by parameter ambiguity.

[0084] S3 Support Set Extraction

[0085] In a power distribution network, the operating time of nonlinear loads is much longer than the interval between two measurement time sections. This ensures that the location and frequency of most harmonic sources remain unchanged between the two time sections. The set of position indices corresponding to all non-zero elements in the sparse vector of the previous moment is extracted as the support set. Using the support set as prior knowledge can effectively decompose the blind optimization in high dimensions, greatly reducing the algorithm's rigid requirement for the number of measurement channels and ensuring the absolute stability of the state estimation along the time axis.

[0086] S3.1 Calculate the local least squares error covariance matrix

[0087] exist At that moment, the system utilizes the set of active harmonic sources at that time. Unbiased least squares tracking was performed; based on Perceptual submatrix after whitening at each time step The covariance matrix of the estimation error of this set of harmonic sources is obtained by analysis:

[0088]

[0089] S3.2 Extract the standard deviation of the target node

[0090] For the set of the th For each potential harmonic source node, calculate the standard deviation of the node state estimation error:

[0091]

[0092] S3.3 Generates an adaptive detection threshold by combining confidence levels.

[0093] In order to The decision of whether a node should continue to be included in the "known harmonic sources" set requires setting a threshold to prevent the inclusion of "pseudo-harmonic sources" in the support set. A statistical confidence multiplier is introduced. Generate a unique node Adaptive hard threshold at time:

[0094]

[0095] in The value can be selected based on the required precision, for example... Under the corresponding normal distribution The confidence probability.

[0096] S3.4 Prior Support Set Generation

[0097] exist During the algorithm initialization phase at time 1, The best estimated amplitude of the entire network at each time step Compare with the threshold calculated above:

[0098]

[0099] Only nodes that satisfy this inequality can be included. Prior support set of time middle.

[0100] S4 Residual Signal Stripping and Burst Source Detection

[0101] S4.1 Active Submatrix Extraction:

[0102] Assuming Step 3.4 has obtained the result containing Support set of elements ,according to From the complete global whitening perception matrix Extract this from The column vectors corresponding to each index are reorganized into an active submatrix. ;

[0103] S4.2 Unbiased Least Squares Estimation

[0104] For known support sets The active harmonic sources within the sensor are directly tracked for high-precision amplitude and phase using the Moore-Penrose pseudo-inverse. This step avoids the amplitude shrinkage to zero bias that can easily occur when using compressed sensing alone.

[0105]

[0106] Step 4.3 Residual Signal Stripping

[0107] By stripping away the measurement responses generated by known harmonic source mappings from the total whitened observation signal, the pure least-squares residuals are obtained. :

[0108]

[0109] Step 4.4 Global Energy Detection of Burst Sources Based on Whitening Residue

[0110] Before calling the high-dimensional optimization algorithm, it is necessary to determine whether the power grid has generated a newly inserted harmonic source;

[0111] The threshold is obtained by looking up a table given a confidence level. ,like If no new burst source is found, skip the subsequent steps and output the result obtained in S4.2. As an estimation result; if ,determination to If a new harmonic source is connected during the time period, it triggers subsequent steps to perform optimization.

[0112] S5 Space-Frequency Joint Residual Compressed Sensing Optimization for Emerging Bursts

[0113] For the residual signal that triggers the alarm, in the remaining space after removing the known set... The optimization is performed in the middle; in order to take advantage of the sparsity of both spatial and frequency dimensions, a joint penalty function based on the mixture norm is constructed to solve the convex optimization problem:

[0114]

[0115] in:

[0116] for Norms force the vast majority of nodes to return to zero across all frequency components, thereby enabling in-depth exploration of "unknown spatial sparsity".

[0117] The norm further forces the surviving non-zero nodes to satisfy "frequency local sparsity".

[0118] To obtain the tolerance boundary based on the rigorous calculation of the chi-square distribution and avoid blind manual parameter tuning; to obtain the preliminary estimated state of the newly added harmonic source. .

[0119] S6 Robust Dual Threshold Detection and Support Set Update

[0120] To prevent overfitting caused by residual optimization, after calculating... The following detection and rejection steps must then be performed:

[0121] S6.1 adds threshold detection:

[0122] because Originating from tolerance of The minimization problem theoretically has a bounded maximum reconstruction error; therefore, It can be strictly defined as a factor of the worst-case error of compressed sensing:

[0123]

[0124] in The value of can be selected according to the confidence level requirements, and can be selected from 3 to 5 according to the normal distribution characteristics;

[0125] Select Medium amplitude greater than Only the node that was identified was formally designated as a "new outbreak source" and added to the temporary support set. .

[0126] S6.2 Combined Unbiased LS Debiasing and Vanishing Source Removal:

[0127] use Re-execute the global unbiased LS estimation.

[0128] calculate :

[0129]

[0130] in The covariance matrix obtained using the temporary support set when re-performing the global unbiased LS estimation;

[0131] If the amplitude of a certain harmonic source in the second LS estimation drops to The following will permanently remove it from the support set and re-execute S6.2 until no amplitude drops to [a certain value]. the following;

[0132] Finally, the robust state vector after two rounds of LS debiasing is used as... The final solution for the entire network space frequency harmonic state estimation at time 1 Output and update the support set. The signal is then transmitted to the next section, forming a perfect closed loop for dynamic tracking.

[0133] Example 2

[0134] like Figure 2 The diagram illustrates the structure of an electronic device 1 that can be used to implement embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0135] like Figure 2 As shown, the electronic device 1 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from storage unit 18. The RAM 13 may also store various programs and data required for the operation of the electronic device 1. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0136] Multiple components in electronic device 1 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of monitors, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 1 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0137] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as harmonic state estimation methods for distribution networks with limited vector measurement and ambiguous parameters.

[0138] In some embodiments, the harmonic state estimation method for distribution networks with limited surface measurement and ambiguous parameters can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 1 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the harmonic state estimation method for distribution networks with limited surface measurement and ambiguous parameters described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the harmonic state estimation method for distribution networks with limited surface measurement and ambiguous parameters by any other suitable means (e.g., by means of firmware).

[0139] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0140] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0141] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0142] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0143] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0144] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0145] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for harmonic state estimation in distribution networks with limited surface vector measurement and ambiguous parameters, characterized in that, Includes the following steps: S1 Model Establishment: Based on the total number of nodes in the distribution network and the harmonic frequencies to be analyzed, a global state vector containing spatial and frequency dimensions is established. The single-frequency sensing sub-matrix is ​​calculated based on Kirchhoff's laws. The block diagonal structure space-frequency joint sensing matrix is ​​constructed through Kronecker product operation. A joint measurement equation considering the deviation of line impedance parameters is established. S2 Key Prior Parameter Acquisition and Spatial Whitening Mapping: Extract the physical parameters of the distribution network lines, calculate the theoretical network matrix corresponding to each harmonic after skin effect correction, construct the comprehensive error covariance matrix in combination with the noise parameters of the measurement equipment, perform eigenvalue decomposition on the comprehensive error covariance matrix to obtain the whitening matrix, and use the whitening matrix to perform linear transformation on the joint measurement equation to obtain the whitened perception matrix and measurement vector. S3 Support Set Extraction: Based on the harmonic state estimation results and the corresponding estimation error covariance matrix of the previous measurement time, calculate the adaptive detection threshold of each potential harmonic source node, filter the node indexes whose estimated amplitudes at the previous time meet the threshold requirements, and form the prior support set for the current time. S4 Residual Signal Stripping and Burst Source Detection: Based on the prior support set, the corresponding column vectors are extracted from the whitened perception matrix to form an active submatrix. The state estimates of the harmonic sources in the prior support set are solved by unbiased least squares operation. The measurement responses corresponding to the state estimates are stripped from the whitened measurement vectors to obtain the residual signal. S5 Space-Frequency Joint Residual Compressed Sensing Optimization for New Burst Sources: Energy threshold detection is performed on the residual signal. If it is determined that there is a new harmonic source, space-frequency joint residual compressed sensing optimization is performed in the index space outside the prior support set using hybrid norm constraints to solve for the preliminary estimated state of the new harmonic source. S6 High Robust Dual Threshold Detection and Support Set Update: Perform new amplification threshold detection on the preliminary estimated state of newly added harmonic sources and add qualified nodes to the temporary support set; Based on the temporary support set, global unbiased least squares estimation is re-executed, and failed harmonic sources are eliminated through amplitude drop detection. The final support set is then updated, and the network-wide harmonic state estimation result at the current moment is output.

2. The method for harmonic state estimation of a distribution network with scarce surface vector measurement and fuzzy parameters according to claim 1, characterized in that, In step S1, the global state vector is reduced to a one-dimensional vector by combining the injected current vectors of all network nodes under each harmonic frequency; the joint measurement equation combines the line parameter perturbation and the measurement equipment noise into an equivalent comprehensive correlation noise, which is used as the error term of the measurement equation.

3. The method for harmonic state estimation in a distribution network with limited surface vector measurement and ambiguous parameters according to claim 1, characterized in that, In step S2, the theoretical network matrix includes the node admittance matrix and current transfer matrix corresponding to each harmonic; the comprehensive error covariance matrix is ​​constructed by combining the prior parameter error variance and the state quantity at the previous moment, and the relative error standard deviation of the prior parameter error variance is 5%.

4. The method for harmonic state estimation in a distribution network with limited surface vector measurement and ambiguous parameters according to claim 1, characterized in that, In step S2, the spatial whitening mapping is used to eliminate the correlation of measurement channels caused by the topological coupling of the distribution network, balance the dimensional differences between voltage measurement and current measurement, and convert the equivalent integrated correlated noise into spatially uncorrelated standard white noise, thereby suppressing the multi-level amplification of errors caused by parameter deviation.

5. The method for harmonic state estimation of a distribution network with scarce surface vector measurement and fuzzy parameters according to claim 1, characterized in that, Step S3 specifically includes: S3.1 Based on the whitened perceptron submatrix and the corresponding support set of the previous time step, the covariance matrix of the harmonic state estimation error of the previous time step is calculated analytically. S3.2 For each potential harmonic source node, extract the standard deviation of its state estimation error, and combine it with the statistical confidence multiplier to generate an adaptive detection threshold specific to that node. S3.3 compares the estimated harmonic amplitude of each node at the previous time step with the corresponding adaptive detection threshold, and includes the node index with amplitude greater than the threshold in the prior support set at the current time step.

6. The method for harmonic state estimation in a distribution network with limited surface vector measurement and ambiguous parameters according to claim 1, characterized in that, In step S4, the unbiased least squares operation is solved using Moore-Penrose pseudo-inverse to correct the magnitude shrinkage to zero bias caused by pure compressed sensing.

7. The method for harmonic state estimation of a distribution network with scarce surface vector measurement and fuzzy parameters according to claim 1, characterized in that, In step S5, the hybrid norm constraint includes Norm and norm, where Norm constraints on the overall sparsity of nodes across the entire frequency band Norm constraints are applied to the local sparsity of non-zero nodes in the frequency dimension; the tolerance boundary of the optimization problem is determined based on the chi-square distribution.

8. The method for harmonic state estimation in a distribution network with limited surface vector measurement and ambiguous parameters according to claim 1, characterized in that, In step S6, the new amplification threshold is a multiple of the worst-case reconstruction error of compressed sensing, and the value of the coefficient is in the range of 3 to 5, corresponding to the confidence level requirement under normal distribution.

9. The method for harmonic state estimation of a distribution network with scarce surface vector measurement and fuzzy parameters according to claim 1, characterized in that, In step S6, the amplitude drop detection process is as follows: after completing the global unbiased least squares estimation based on the temporary support set, if the estimated amplitude of the original harmonic source drops below its corresponding adaptive detection threshold, the node is removed from the support set, and the unbiased least squares estimation and amplitude drop detection are repeated until no node is removed.

10. The method for harmonic state estimation of a distribution network with scarce surface vector measurement and fuzzy parameters according to claim 1, characterized in that, The final support set obtained in step S6 will be passed to the next measurement time section as the basis for calculating the a priori support set at the next moment, forming a dynamic continuous tracking closed loop of the distribution network harmonic state.

11. An electronic device, characterized in that, The device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the operation as described in claim 1.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to perform the functions described in claim 1.

Images