Adaptive filter-based distributed power grid power quality optimization method and device

Abstract

The application belongs to the technical field of power quality control, and particularly relates to a distributed power grid power quality optimization method and equipment based on adaptive filtering, which realizes distributed power grid power quality optimization through the steps of power quality parameter collection, disturbance identification and classification, adaptive filtering algorithm construction and parameter optimization, distributed collaborative control strategy execution and feedback regulation; precise disturbance identification is realized through an improved wavelet packet transform and an optimized support vector machine, an adaptive filtering model with a variable step factor is constructed to balance filtering accuracy and response speed, and combined with the distributed collaborative control strategy, the instruction of each node is collaboratively optimized; the application solves the problems of poor adaptability, slow response and low accuracy of existing methods, can quickly and accurately suppress voltage sag, harmonic pollution and other disturbances, improves the power quality and operation stability of the distributed power grid, ensures the normal operation of sensitive loads, and reduces the failure rate of the power grid.

Description

Technical Field

[0001] This invention belongs to the field of power quality control technology, specifically relating to a method and device for optimizing power quality in distributed power grids based on adaptive filtering. Background Technology

[0002] With the large-scale integration of distributed power sources (such as photovoltaic and wind power) and the widespread application of power electronic equipment, power quality problems in power grids are becoming increasingly prominent, mainly manifested as voltage sags, harmonic pollution, voltage fluctuations, and flicker. These problems not only affect the normal operation of sensitive power loads (such as precision instruments and industrial automation equipment), but may also reduce the operating efficiency of the power grid, accelerate the aging of power equipment, and even trigger power grid failures.

[0003] Existing power quality optimization methods mainly include passive filtering, active power filtering, and voltage compensation based on traditional control algorithms. Among them, passive filtering has a simple structure and low cost, but its filtering characteristics are fixed, making it difficult to adapt to dynamic changes in grid operating conditions, and its suppression effect on dynamic harmonics and random voltage fluctuations is poor. Active power filtering has dynamic compensation capabilities, but the fixed parameter control algorithms (such as PI control) used in traditional active filtering have problems such as slow response speed and low compensation accuracy in scenarios with sudden changes in grid parameters and drastic load fluctuations, making it difficult to achieve accurate tracking and compensation for power quality disturbances.

[0004] Furthermore, in distributed power grids, the load distribution of each node is dispersed and the operating conditions are complex and variable. Traditional centralized power quality optimization methods suffer from large communication delays and lagging control command execution, further reducing the effectiveness of power quality optimization. Therefore, there is an urgent need for a distributed power grid power quality optimization method with adaptive capabilities, fast response speed, and high compensation accuracy to address the shortcomings of existing technologies. Summary of the Invention

[0005] To address the problems of poor adaptability, slow response speed, and low compensation accuracy in existing distributed power grid power quality optimization methods, this invention provides a distributed power grid power quality optimization method and device based on adaptive filtering. By constructing an adaptive filtering algorithm and a distributed cooperative control strategy, it achieves accurate and rapid suppression of power quality disturbances in the power grid, thereby improving the power quality and operational stability of the distributed power grid.

[0006] This invention is implemented using the following technical solution: a distributed power grid power quality optimization method based on adaptive filtering, comprising: The initial power quality parameters of each node in the distributed power grid are collected in real time, and the initial power quality parameters are preprocessed to obtain standardized power quality parameters. Based on power quality parameters, an improved wavelet packet transform algorithm is used to extract time-frequency domain feature parameters of power quality disturbance signals. An optimized support vector machine classification model is then used to classify and identify the time-frequency domain feature parameters of power quality disturbance signals, thereby identifying the disturbance type of power quality disturbance signals. For power quality disturbance signals of different disturbance types, an adaptive filtering model based on minimum mean square error is constructed, and the filtering parameter optimization strategy of the adaptive filtering model is adaptively adjusted by a variable step size factor optimization algorithm. Each node of the distributed power grid receives and analyzes power quality disturbance signals and optimized filtering parameters through a distributed communication network, obtains filtering adjustment instructions from the active filtering device of the corresponding node, and the active filtering device of each node of the distributed power grid performs adjustment and optimization operations according to the corresponding filtering adjustment instructions.

[0007] Preferably, the parameter types of the initial power quality parameters include three-phase voltage signals, three-phase current signals, and frequency signals; The initial power quality parameters are preprocessed to obtain standardized power quality parameters, including: An instrumentation amplifier is used to adaptively amplify the initial power quality parameters; A first-order RC low-pass filter and a Kalman filter algorithm are used to perform secondary noise reduction on the initial power quality parameters; GPS-synchronized clocks are used to enable each node in the distributed power grid to synchronously sample initial power quality parameters.

[0008] Preferably, based on power quality parameters, an improved wavelet packet transform algorithm is used to extract time-frequency domain feature parameters of the power quality disturbance signal, including: An adaptive threshold function is introduced to perform threshold processing on the wavelet packet decomposition coefficients of the wavelet packet transform algorithm, resulting in an improved wavelet packet transform algorithm. The power quality parameters are feature-extracted using an improved wavelet packet transform algorithm to obtain the time-frequency domain feature parameters of the power quality disturbance signal; among which, The time-domain characteristic parameters of power quality disturbance signals include peak value, amplitude fluctuation range, and duration; the frequency-domain characteristic parameters of power quality disturbance signals include harmonic order, harmonic amplitude, and total harmonic distortion rate.

[0009] Preferably, the optimized support vector machine classification model is used to classify and identify the time-frequency domain feature parameters of the power quality disturbance signal, and to identify the disturbance type of the power quality disturbance signal, including: A support vector machine (SVM) classification model is constructed, which employs a radial basis function (RBF) kernel. The penalty factor and kernel function parameters of the support vector machine classification model are optimized by using the particle swarm optimization algorithm to obtain the optimized support vector machine classification model. The optimized support vector machine classification model is used to classify and identify power quality disturbance signals, thus identifying the disturbance type. Types of power quality disturbances include voltage sags, harmonic pollution, and voltage fluctuations.

[0010] Preferably, for power quality disturbance signals of different disturbance types, an adaptive filtering model based on minimum mean square error is constructed, and the filtering parameter optimization strategy of the adaptive filtering model is adaptively adjusted through a variable step size factor optimization algorithm, including: An adaptive filtering model based on minimum mean square error (LMS) is constructed, and a variable step size factor optimization algorithm is introduced; the step size factor adjustment formula of the variable step size factor optimization algorithm is expressed as: μ(k+1)=μ(k)×exp(-0.05×|e(k)|) Where μ(k) is the step size factor of the kth iteration, α is the attenuation coefficient, and e(k) is the filtering error signal of the kth iteration; Based on the identified power quality disturbance signal, the value of the initial step size factor is adjusted, thereby adjusting the filter parameter optimization strategy of the adaptive filtering model.

[0011] Preferably, each node of the distributed power grid receives and parses power quality disturbance signals and optimized filtering parameters through a distributed communication network, obtains filtering adjustment instructions for the corresponding active filter device, and the active filter device of each node of the distributed power grid performs adjustment and optimization operations according to the corresponding filtering adjustment instructions, including: Each node of the distributed power grid interacts with information via industrial Ethernet, receiving power quality disturbance signals and optimized filtering parameters. Each node in the distributed power grid predicts the trend of power quality disturbance changes over the next 1-3 sampling periods based on local power quality signals and feedback information from neighboring nodes. By minimizing the sum of squared errors between the power quality parameters and the standard values, the filtering adjustment command for the active filter device at the corresponding node is obtained, and the adjustment and optimization operation is performed according to the filtering adjustment command.

[0012] Preferably, the active filter device at each node of the distributed power grid outputs compensation current or compensation voltage in real time according to the filter adjustment command; at the same time, the local control unit at each node of the distributed power grid collects the optimized power quality parameters in real time and compares them with the standard power optimization parameter threshold. If the optimized power quality parameters collected in real time are all greater than the standard power optimization parameter threshold, the error signal is fed back to the adaptive filter model to readjust the filter parameters and compensation command until the power quality parameters meet the national standard requirements.

[0013] The present invention also provides a distributed power quality optimization device for power grids based on adaptive filtering, comprising: The data processing module is used to collect the initial power quality parameters of each node of the distributed power grid in real time, and to preprocess the initial power quality parameters to obtain standardized power quality parameters. The disturbance type identification module is used to extract the time-frequency domain feature parameters of the power quality disturbance signal based on the power quality parameters and using an improved wavelet packet transform algorithm. It then uses an optimized support vector machine classification model to classify and identify the time-frequency domain feature parameters of the power quality disturbance signal, thereby identifying the disturbance type of the power quality disturbance signal. The filter parameter optimization module is used to construct an adaptive filter model based on minimum mean square error for power quality disturbance signals of different disturbance types, and to adaptively adjust the filter parameter optimization strategy of the adaptive filter model through a variable step size factor optimization algorithm. The adjustment and optimization module is used to control each node of the distributed power grid to receive and parse power quality disturbance signals and optimized filtering parameters through a distributed communication network, obtain the filtering adjustment instructions of the active filter device of the corresponding node, and control the active filter device of each node of the distributed power grid to perform adjustment and optimization operations according to the corresponding filtering adjustment instructions.

[0014] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the distributed power quality optimization method for adaptive filtering based on the aforementioned technical solution.

[0015] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the distributed power quality optimization method for adaptive filtering as described in the foregoing technical solution.

[0016] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a method and device for optimizing power quality in a distributed power grid based on adaptive filtering. The method achieves power quality optimization through power quality parameter acquisition, disturbance identification and classification, adaptive filtering algorithm construction and parameter optimization, distributed collaborative control strategy execution, and feedback adjustment steps. It achieves accurate disturbance identification through improved wavelet packet transform and optimized support vector machine, constructs an adaptive filtering model with a variable step size factor to balance filtering accuracy and response speed, and combines distributed collaborative control strategy to achieve collaborative optimization of instructions from each node. This invention solves the problems of poor adaptive capability, slow response, and low accuracy in existing methods, and can quickly and accurately suppress disturbances such as voltage sags and harmonic pollution, improve the power quality and operational stability of the distributed power grid, ensure the normal operation of sensitive loads, and reduce the occurrence rate of power grid faults. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a distributed power quality optimization method for power grids based on adaptive filtering, provided by the present invention.

[0019] Figure 2 This is a logical schematic diagram of a distributed power quality optimization method for power grids based on adaptive filtering provided by the present invention.

[0020] Figure 3 This is a schematic diagram of the overall architecture of the distributed power quality optimization system in the distributed power quality optimization method based on adaptive filtering provided by the present invention.

[0021] Figure 4 This is a schematic diagram of the structure of a distributed power grid power quality optimization device based on adaptive filtering provided by the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] It should be noted that the structures, proportions, sizes, etc., shown in the accompanying drawings of this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed in the specification, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportional relationships, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should fall within the scope of the technical content disclosed in the present invention. It should be noted that in this specification, relational terms such as "first" and "second" are only used to distinguish one entity from several other entities, and do not necessarily require or imply any actual relationship or order between these entities.

[0024] This invention provides an embodiment: such as Figure 1 and Figure 2 As shown, this invention provides a distributed power quality optimization method for power grids based on adaptive filtering, comprising: S110: Real-time acquisition of initial power quality parameters of each node in the distributed power grid, and preprocessing of the initial power quality parameters to obtain standardized power quality parameters.

[0025] The overall architecture of the distributed power grid power quality optimization system involved in this invention is as follows: Figure 3 As shown. Figure 3 In the diagram, 1 represents a distributed grid node, 2 represents a voltage sensor, 3 represents a current sensor, 4 represents a local control unit, 5 represents an active filter, 6 represents a distributed communication network, 7 represents a sensitive load, and 8 represents a distributed power source. The components interact via the distributed communication network 6. The local control unit 4 is electrically connected to sensors (2, 3) and the active filter 5, respectively, to receive signals and output control commands.

[0026] In an embodiment of the present invention, voltage sensors 2 and current sensors 3 are deployed at each node of the distributed power grid to collect the three-phase voltage signal, three-phase current signal, and frequency signal of each node in real time, which are used as initial power quality parameters. After collecting the initial power quality parameters, data preprocessing is performed on the collected initial power quality parameters. The specific preprocessing process includes: An instrumentation amplifier is used to adaptively amplify the initial power quality parameters; A first-order RC low-pass filter and a Kalman filter algorithm are used to perform secondary noise reduction on the initial power quality parameters; GPS-synchronized clocks are used to enable each node in the distributed power grid to synchronously sample initial power quality parameters.

[0027] After preprocessing, the initial power quality parameters are converted into standardized power quality parameters.

[0028] S120: Based on power quality parameters, an improved wavelet packet transform algorithm is used to extract the time-frequency domain feature parameters of the power quality disturbance signal. The optimized support vector machine classification model is then used to classify and identify the time-frequency domain feature parameters of the power quality disturbance signal, thereby identifying the disturbance type of the power quality disturbance signal.

[0029] In this embodiment, the wavelet packet transform algorithm is improved by introducing an adaptive threshold function. The improved wavelet packet transform algorithm is used to extract the time-frequency domain feature parameters of the power quality disturbance signal based on the standardized power quality parameters.

[0030] Specifically, the time-domain characteristic parameters of power quality disturbance signals include peak value, amplitude fluctuation range, and duration; the frequency-domain characteristic parameters of power quality disturbance signals include harmonic order, harmonic amplitude, and total harmonic distortion rate.

[0031] Specifically, an improved wavelet packet transform algorithm is used to extract features from the standardized power quality parameters. The wavelet basis is selected as db4 wavelet, the decomposition level is 4, and an adaptive threshold function is introduced, with threshold λ=σ×√(2lnN), where σ is the noise standard deviation and N is the signal length. Thresholding is performed on the wavelet packet decomposition coefficients to extract the time-domain and frequency-domain feature parameters of the power quality disturbance signal.

[0032] After extracting the time-frequency domain feature parameters of the power quality disturbance signal, the optimized support vector machine classification model is used to classify and identify the time-frequency domain feature parameters of the power quality disturbance signal, thereby identifying the disturbance type of the power quality disturbance signal.

[0033] Specifically, a support vector machine (SVM) classification model is constructed, which uses a radial basis function (RBF) kernel. The penalty factor and kernel function parameters of the support vector machine classification model are optimized by using the particle swarm optimization algorithm to obtain the optimized support vector machine classification model. The optimized support vector machine classification model is used to classify and identify power quality disturbance signals, thus identifying the disturbance type. The types of power quality disturbance signals include voltage sags, harmonic pollution, and voltage fluctuations. Specifically, when the voltage amplitude drops by 30% to 60%, the disturbance type is identified as a voltage sag; when the 3rd, 5th, or 7th harmonics are present, the disturbance type is identified as harmonic pollution; and when the voltage fluctuation range is within 5% to 10%, the disturbance type is identified as voltage fluctuation. In other embodiments of the present invention, the disturbance types of power quality disturbance signals can also be divided into two types: voltage sags and voltage fluctuations, as described above, are categorized as dynamic disturbances. Therefore, the two types of power quality disturbance signals are dynamic disturbances and harmonic pollution.

[0034] S130: For power quality disturbance signals of different disturbance types, an adaptive filtering model based on minimum mean square error is constructed, and the filtering parameter optimization strategy of the adaptive filtering model is adaptively adjusted through a variable step size factor optimization algorithm.

[0035] For power quality disturbances of different types, an adaptive filtering model based on minimum mean square error (LMS) is constructed, and a variable step size factor optimization algorithm is introduced to improve the traditional LMS algorithm. According to the disturbance type identified in step S120, the step size factor variation strategy of the initial parameters of the filtering model is adaptively adjusted: for harmonic pollution disturbances, a smaller initial step size is used to ensure filtering accuracy; for dynamic disturbances such as voltage sags and voltage fluctuations, a larger initial step size is used to improve the response speed. At the same time, the step size is adjusted in real time through the adaptive step size factor adjustment formula to achieve a dynamic balance between filtering accuracy and response speed.

[0036] Specifically, an adaptive filtering model based on minimum mean square error (LMS) is constructed, and a variable step size factor optimization algorithm is introduced; the step size factor adjustment formula of the variable step size factor optimization algorithm is expressed as: μ(k+1)=μ(k)×exp(-0.05×|e(k)|) Where μ(k) is the step size factor of the kth iteration, α is the attenuation coefficient, and e(k) is the filtering error signal of the kth iteration; the initial step size factor μ(0) is set according to the disturbance type, the value of harmonic pollution disturbance μ(0) is 0.01~0.05, and the value of dynamic disturbance μ(0) is 0.1~0.2.

[0037] Based on the identified power quality disturbance signal, the value of the initial step size factor is adjusted, thereby adjusting the filter parameter optimization strategy of the adaptive filtering model.

[0038] In a specific embodiment, α=0.05 is the attenuation coefficient, and e(k) is the filtering error signal of the kth iteration. The initial step size factor μ(0) is set according to the identified disturbance type: for harmonic pollution disturbances, μ(0)=0.03; for dynamic disturbances such as voltage sags and voltage fluctuations, μ(0)=0.15. The step size is adjusted in real time through the above optimization algorithm to achieve a dynamic balance between filtering accuracy and response speed.

[0039] Specifically, the filter parameter optimization strategy for adjusting the initial step size factor for different disturbance types is based on setting differentiated initial step sizes according to the characteristics of the disturbances, and dynamically iteratively adjusting them through a variable step size factor algorithm. This is combined with parameter adaptation of the filter model to achieve a dynamic balance between filtering accuracy and response speed. The specific optimization strategy is as follows: I. Core Design Principles of Optimization Strategies The filtering parameter optimization strategy of this invention is designed around the dynamic characteristics of the disturbance: 1. Harmonic pollution is a steady-state / quasi-steady-state disturbance, which requires high filtering accuracy but relatively low response speed. Therefore, an optimization strategy with a small initial step size is adopted to prioritize minimizing the filtering error. 2. Voltage sags and voltage fluctuations are dynamic transient disturbances. The grid parameters change rapidly and the disturbance duration is short, which requires high response speed. Therefore, an optimization strategy with a large initial step size is adopted to prioritize the rapid tracking of disturbances and the completion of compensation, and then the accuracy is fine-tuned through a variable step size algorithm.

[0040] All initial step size settings are linked to the variable step size factor optimization algorithm. The initial step size provides a basic value for algorithm iteration, and the step size is dynamically adjusted in real time based on the filtering error to ultimately achieve a balance between accuracy and speed.

[0041] The invention clearly categorizes disturbances into harmonic pollution, voltage sag, and voltage fluctuation (the latter two being classified as dynamic disturbances). For each type, a specific optimization strategy is developed, including initial step size settings, iteration rules, and filter model adaptation, as detailed below: (I) Parameter optimization strategy for harmonic pollution disturbance 1. Initial step size factor value: μ(0)∈[0.01,0.05], the preferred value of the invention is 0.03, which is set for small step size; 2. Core optimization objective: Minimize filtering error, ensure the accuracy of harmonic suppression, and reduce the total harmonic distortion rate to below the national standard (≤5%); 3. Iterative adjustment rules: Based on the step size factor formula μ(k+1)=μ(k)×exp (-0.05×|e (k)|), the iteration is based on the small initial step size. The step size decays / increases slowly during the iteration process, and the weight update of the filtering model is smooth, avoiding filtering oscillations caused by the large step size, and ensuring accurate suppression of the 3rd, 5th and 7th characteristic harmonics. 4. Matching parameter adaptation: The filtering order of the adaptive filtering model is set to a high order (such as 16 / 32 order) to improve the resolution of harmonic frequencies and achieve high-precision filtering in conjunction with the small step size strategy.

[0042] S140: Each node of the distributed power grid receives and parses power quality disturbance signals and optimized filtering parameters through a distributed communication network, obtains the filtering adjustment instructions of the active filter device of the corresponding node, and the active filter device of each node of the distributed power grid performs adjustment and optimization operations according to the corresponding filtering adjustment instructions.

[0043] Each node of the distributed power grid interacts with information via industrial Ethernet, receiving power quality disturbance signals and optimized filtering parameters. Each node in a distributed power grid predicts the trend of power quality disturbance changes over the next 1-3 sampling periods based on local power quality signals and feedback information from neighboring nodes. By minimizing the sum of squared errors between the power quality parameters and the standard values, the filtering adjustment command for the active filter device at the corresponding node is obtained, and the adjustment and optimization operation is performed according to the filtering adjustment command.

[0044] Specifically, each node's local control unit interacts with other nodes via industrial Ethernet (communication rate 100Mbps) to share power quality disturbance information and filtering parameters. A distributed model predictive control strategy is adopted to achieve consistency optimization. Each node constructs a local predictive model based on local power quality information and feedback information from neighboring nodes to predict the power quality disturbance change trend for the next two sampling periods. The objective function is set as the sum of squared errors between the power quality parameters and the standard values ​​of national standards (GB / T 12325-2022 "Power Quality - Supply Voltage Deviation" and GB / T 14549-1993 "Power Quality - Harmonics in Public Power Grids"). Local control commands are obtained by minimizing the objective function. The node communication weights are dynamically adjusted based on the electrical distance between nodes and the quality of the communication link. The electrical distance is calculated using impedance, and the communication link quality is evaluated using the packet loss rate. The weight is 0.8~1.0 when the packet loss rate is below 5%, 0.5~0.8 when the packet loss rate is 5%~10%, and 0.1~0.5 when the packet loss rate is above 10%.

[0045] Each node's active filter (using a three-level topology with IGBTs as the switching device) outputs compensation current or adjusts voltage in real time according to the compensation command generated in step 4. Simultaneously, the local control unit collects the optimized power quality parameters in real time and compares them with national standards. If the total harmonic distortion rate is greater than 5%, the voltage deviation is greater than ±7%, or the voltage fluctuation amplitude is greater than 2.5%, the error signal is fed back to the adaptive filtering model to readjust the filtering parameters and compensation commands until the power quality parameters meet the requirements of national standards.

[0046] Tests have shown that, by adopting the method of this invention, the total harmonic distortion rate of each node in the distributed power grid is reduced to below 3%, the voltage sag recovery time is shortened to within 5ms, the voltage fluctuation amplitude is controlled to below 2%, the power quality is significantly improved, the operational stability of sensitive loads is enhanced, and the power grid fault rate is reduced by more than 40%.

[0047] like Figure 4 As shown, the present invention also provides a distributed power grid power quality optimization device based on adaptive filtering, comprising: The data processing module 410 is used to collect the initial power quality parameters of each node of the distributed power grid in real time, and to preprocess the initial power quality parameters to obtain standardized power quality parameters. The disturbance type identification module 420 is used to extract the time-frequency domain feature parameters of the power quality disturbance signal based on the power quality parameters and using an improved wavelet packet transform algorithm. The optimized support vector machine classification model is then used to classify and identify the time-frequency domain feature parameters of the power quality disturbance signal to identify the disturbance type of the power quality disturbance signal. The filter parameter optimization module 430 is used to construct an adaptive filter model based on minimum mean square error for power quality disturbance signals of different disturbance types, and to adaptively adjust the filter parameter optimization strategy of the adaptive filter model through a variable step size factor optimization algorithm. The adjustment and optimization module 440 is used to control each node of the distributed power grid to receive and parse power quality disturbance signals and optimized filtering parameters through a distributed communication network, obtain the filtering adjustment instructions of the active filter device of the corresponding node, and control the active filter device of each node of the distributed power grid to perform adjustment and optimization operations according to the corresponding filtering adjustment instructions.

[0048] To implement the embodiments, the present invention also proposes a computer device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the adaptive matching method for process parameters of an automatic anchor cable drill frame for coal mines as described in the foregoing technical solutions.

[0049] The present invention also proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the adaptive matching method for process parameters of an automatic anchor cable drill frame for coal mines as described in the foregoing technical solution.

[0050] Although embodiments of the present invention have been shown and described above, it is understood that the 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 embodiments within the scope of the present invention.

Claims

1. A distributed power quality optimization method for power grids based on adaptive filtering, characterized in that, include: The initial power quality parameters of each node in the distributed power grid are collected in real time, and the initial power quality parameters are preprocessed to obtain standardized power quality parameters. Based on the power quality parameters, an improved wavelet packet transform algorithm is used to extract the time-frequency domain feature parameters of the power quality disturbance signal. The optimized support vector machine classification model is then used to classify and identify the time-frequency domain feature parameters of the power quality disturbance signal, thereby identifying the disturbance type of the power quality disturbance signal. For power quality disturbance signals of different disturbance types, an adaptive filtering model based on minimum mean square error is constructed, and the filtering parameter optimization strategy of the adaptive filtering model is adaptively adjusted by a variable step size factor optimization algorithm. Each node of the distributed power grid receives and analyzes the power quality disturbance signal and optimized filtering parameters through a distributed communication network, obtains the filtering adjustment instructions of the active filter device of the corresponding node, and the active filter device of each node of the distributed power grid performs adjustment and optimization operations according to the corresponding filtering adjustment instructions.

2. The distributed power quality optimization method for power grids based on adaptive filtering according to claim 1, characterized in that, The initial power quality parameters include three-phase voltage signals, three-phase current signals, and frequency signals. The initial power quality parameters are preprocessed to obtain standardized power quality parameters, including: The initial power quality parameters are adaptively amplified using an instrumentation amplifier. The initial power quality parameters are denoised a second time using a first-order RC low-pass filter and a Kalman filter algorithm. GPS-synchronized clocks are used to enable each node of the distributed power grid to synchronously sample the initial power quality parameters.

3. The distributed power quality optimization method for power grids based on adaptive filtering according to claim 1, characterized in that, Based on the power quality parameters, an improved wavelet packet transform algorithm is used to extract the time-frequency domain feature parameters of the power quality disturbance signal, including: An adaptive threshold function is introduced to perform threshold processing on the wavelet packet decomposition coefficients of the wavelet packet transform algorithm, resulting in an improved wavelet packet transform algorithm. The power quality parameters are feature-extracted using an improved wavelet packet transform algorithm to obtain the time-frequency domain feature parameters of the power quality disturbance signal; wherein, The time-domain characteristic parameters of power quality disturbance signals include peak value, amplitude fluctuation range, and duration; the frequency-domain characteristic parameters of power quality disturbance signals include harmonic order, harmonic amplitude, and total harmonic distortion rate.

4. The distributed power quality optimization method for power grids based on adaptive filtering according to claim 3, characterized in that, An optimized support vector machine classification model is used to classify and identify the time-frequency domain feature parameters of power quality disturbance signals, and to identify the disturbance types of power quality disturbance signals, including: Construct a support vector machine classification model, wherein the support vector machine classification model adopts a radial basis kernel function; The penalty factor and kernel function parameters of the support vector machine classification model are optimized by particle swarm optimization algorithm to obtain the optimized support vector machine classification model; The optimized support vector machine classification model is used to classify and identify power quality disturbance signals, thus identifying the disturbance type. Types of power quality disturbances include voltage sags, harmonic pollution, and voltage fluctuations.

5. The distributed power quality optimization method for power grids based on adaptive filtering according to claim 1, characterized in that, For power quality disturbance signals of different disturbance types, an adaptive filtering model based on minimum mean square error is constructed. The optimization strategy for the filtering parameters of the adaptive filtering model is adaptively adjusted using a variable step-size factor optimization algorithm, including: An adaptive filtering model based on minimum mean square error (LMS) is constructed, and a variable step size factor optimization algorithm is introduced; wherein, the step size factor adjustment formula of the variable step size factor optimization algorithm is expressed as: μ(k+1)=μ(k)×exp(-0.05×|e(k)|) Where μ(k) is the step size factor of the kth iteration, α is the attenuation coefficient, and e(k) is the filtering error signal of the kth iteration; Based on the disturbance type of the identified power quality disturbance signal, the value of the initial step size factor is adjusted, thereby adjusting the filter parameter optimization strategy of the adaptive filtering model.

6. The distributed power quality optimization method for power grids based on adaptive filtering according to claim 1, characterized in that, Each node of the distributed power grid receives and analyzes the power quality disturbance signal and optimized filtering parameters through a distributed communication network, obtains the filtering adjustment command of the corresponding node's active filter, and the active filter of each node performs adjustment and optimization operations according to the corresponding filtering adjustment command, including: Each node of the distributed power grid interacts with information via industrial Ethernet, receiving power quality disturbance signals and optimized filtering parameters. Each node in a distributed power grid predicts the trend of power quality disturbance changes over the next 1-3 sampling periods based on local power quality signals and feedback information from neighboring nodes. By minimizing the sum of squared errors between the power quality parameters and the standard values, the filtering adjustment command of the active filter device at the corresponding node is obtained, and the adjustment and optimization operation is performed according to the filtering adjustment command.

7. A distributed power quality optimization method for power grids based on adaptive filtering according to claim 6, characterized in that, The active filter device at each node of the distributed power grid outputs compensation current or compensation voltage in real time according to the filter adjustment command. At the same time, the local control unit at each node of the distributed power grid collects the optimized power quality parameters in real time and compares them with the standard power optimization parameter threshold. If the optimized power quality parameters collected in real time are all greater than the standard power optimization parameter threshold, the error signal is fed back to the adaptive filter model to readjust the filter parameters and compensation command until the power quality parameters meet the national standard requirements.

8. A distributed power grid power quality optimization device based on adaptive filtering, characterized in that, include: The data processing module is used to collect the initial power quality parameters of each node of the distributed power grid in real time, and to preprocess the initial power quality parameters to obtain standardized power quality parameters. The disturbance type identification module is used to extract the time-frequency domain feature parameters of the power quality disturbance signal based on the power quality parameters, using an improved wavelet packet transform algorithm, and then use an optimized support vector machine classification model to classify and identify the time-frequency domain feature parameters of the power quality disturbance signal to identify the disturbance type of the power quality disturbance signal. The filter parameter optimization module is used to construct an adaptive filter model based on minimum mean square error for power quality disturbance signals of different disturbance types, and to adaptively adjust the filter parameter optimization strategy of the adaptive filter model through a variable step size factor optimization algorithm. The adjustment and optimization module is used to control each node of the distributed power grid to receive and parse the power quality disturbance signal and the optimized filtering parameters through the distributed communication network, obtain the filtering adjustment instructions of the active filter device of the corresponding node, and control the active filter device of each node of the distributed power grid to perform adjustment and optimization operations according to the corresponding filtering adjustment instructions.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that, The processor executes the computer program to implement the distributed power quality optimization method for power grids based on adaptive filtering as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the distributed power quality optimization method for power grids based on adaptive filtering as described in any one of claims 1-7.

Images