Method and device for controlling power quality of AC / DC power distribution network

Abstract

The invention discloses an AC / DC power distribution network power quality control method and device, and the method comprises the steps: collecting power quality data in an AC / DC power distribution network, carrying out the feature analysis of the power quality data, and obtaining a current feature parameter and a voltage abnormality identifier; carrying out differential compression processing on the power quality data based on the current characteristic parameters and the voltage abnormity identification; transmitting the compressed data and feature information through a redundant channel comprising a main communication link and a standby communication link; and generating electric energy quality alarm information according to the transmitted data, and performing power grid operation decision control according to the information. According to the method, the electric energy quality data of the plurality of Internet of Things electric energy quality sensors in the AC / DC power distribution network are collected, and harmonic analysis is performed on the current waveform of the electric energy quality data, so that the accuracy and robustness of feature extraction are ensured; therefore, the requirements of the modern AC / DC hybrid power distribution network for high-precision, high-reliability and quick-response power quality control can be met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of quality control, and particularly relates to a method and equipment for controlling power quality of an AC-DC distribution network. BACKGROUND

[0002] With large-scale access of distributed power sources, energy storage devices and flexible interconnected devices, the AC-DC hybrid distribution network, while providing efficient and reliable power supply, also faces increasingly serious power quality problems such as harmonic pollution, voltage fluctuation, frequency deviation and transient impact. In order to ensure power quality, it is usually necessary to monitor and control the voltage, current and power of each node of the distribution network in real time.

[0003] In the prior art, an online power quality monitoring terminal can collect voltage and current waveforms on a bus to realize local monitoring. However, due to the limited monitoring nodes, the data is mostly local measurement, which is difficult to accurately reflect the overall power quality status of the entire AC-DC distribution network, and there is a problem of insufficient data coverage.

[0004] To improve the monitoring coverage, some technical solutions introduce an Internet of Things (IoT) sensor network to collect power grid operation data through a large number of low-cost sensor nodes and transmit the data through a wireless network to achieve a wider power quality perception. However, the IoT network is generally based on low-power wide-area network (LPWAN) and cellular network communication technologies, which have problems such as limited bandwidth, time delay fluctuation and data packet loss. In particular, in the dynamic operation scenario of the AC-DC distribution network, millisecond-level synchronization cannot be guaranteed, which may result in power quality voltage anomalies such as voltage sag and harmonic surge not being captured in time, affecting fast control and dynamic adjustment.

[0005] Therefore, the existing power quality control technology still has deficiencies in monitoring coverage, real-time performance and reliability, and it is difficult to meet the demand of modern AC-DC hybrid distribution networks for high-precision, high-reliability and fast-response power quality control. SUMMARY

[0006] The present application provides a method and equipment for controlling power quality of an AC-DC distribution network, which mainly aims to solve the problem that the existing power quality control technology still has deficiencies in monitoring coverage, real-time performance and reliability, and it is difficult to meet the demand of modern AC-DC hybrid distribution networks for high-precision, high-reliability and fast-response power quality control.

[0007] In a first aspect, to achieve the above object, the present application provides a method for controlling power quality of an AC-DC distribution network, comprising: The power quality data in the AC-DC distribution network is collected, and the power quality data is analyzed to obtain current characteristic parameters and voltage abnormality identification; the power quality data is processed by differential compression based on the current characteristic parameters and the voltage abnormality identification; the compressed data and characteristic information are transmitted through a redundant channel comprising a main communication link and a backup communication link; power quality alarm information is generated according to the transmitted data, and grid operation decision control is performed according to the information.

[0008] In the present disclosure, the feature analysis includes: harmonic analysis of the current waveform to extract harmonic characteristic parameters; adaptive threshold analysis of voltage fluctuation to generate voltage abnormality identification.

[0009] In the present disclosure, the differential compression processing includes: based on the current characteristic parameters and the voltage abnormality identification, abnormal state judgment is performed on each data frame in the power quality data; according to the judgment result, a lossy compression strategy is used for normal data frames, and a lossless compression strategy is used for abnormal data frames, and the compressed data is output.

[0010] In the present disclosure, the current waveform is analyzed to extract harmonic characteristic parameters, specifically: the power quality data is preprocessed to obtain a standardized current waveform signal; the standardized current waveform signal is processed by frame to obtain an analysis frame signal; the analysis frame signal is analyzed by frequency spectrum to obtain the amplitude and phase parameters of the fundamental frequency and its harmonic frequency; the content rate of each harmonic and the total harmonic distortion rate are calculated based on the amplitude and phase parameters.

[0011] In the present disclosure, the adaptive threshold analysis of voltage fluctuation to generate voltage abnormality identification, specifically: obtain the historical voltage waveform, calculate the historical voltage mean and historical voltage standard deviation; real-time acquisition of voltage waveform and extraction of instantaneous voltage value; calculate the fluctuation amplitude according to the historical voltage mean and the instantaneous voltage value; based on the historical voltage mean, the historical voltage standard deviation and the preset fluctuation factor, a dynamic fluctuation threshold range is generated; determine whether the fluctuation amplitude is within the dynamic fluctuation threshold range; if it is within the range, it is determined as a normal voltage waveform; if it is out of range, it is determined as a voltage abnormal waveform and a voltage abnormal label is generated.

[0012] In the present disclosure, the differential compression processing of the power quality data includes: judging whether a target frame in the power quality data is marked as voltage abnormality or harmonic abnormality frame by frame, the harmonic abnormality is obtained by comparing the amplitude value in the current harmonic feature with the preset amplitude threshold; if the target frame has no voltage abnormality and harmonic abnormality, it is marked as a normal frame, differential encoding and lossy compression are performed to obtain a compressed normal waveform; if the target frame has any abnormality, it is marked as an abnormal frame, lossless compression is performed to obtain a compressed abnormal waveform; all compressed normal waveforms and compressed abnormal waveforms are summarized to generate a compressed waveform.

[0013] The compression abnormal waveform of the present disclosure, the specific acquisition method is: wavelet transform is carried out on the abnormal frame power quality data, and a set of wavelet coefficients of multiple scales is obtained; based on the energy distribution of the set of wavelet coefficients, a set of significant coefficients with significant energy is screened; the set of significant coefficients is quantized and losslessly encoded to obtain a set of encoding coefficients; the encoding coefficients are inversely transformed to generate a sparse approximate waveform, and a residual signal between the original waveform and the sparse approximate waveform is calculated; the residual signal is losslessly compressed to obtain a compressed residual; the encoding coefficients and the compressed residual are combined as a compressed abnormal waveform.

[0014] The present disclosure, the power quality alarm information is generated according to the transmitted data, and the grid operation decision control is carried out according to the information, comprising: the current characteristic parameters, the voltage abnormality identification and the compression waveform are weighted and fused to generate a fusion data set; the fusion data set is formatted and packaged, and a compression fingerprint code generated based on the compression waveform is embedded to obtain an update data set, an alarm data packet containing the update data set and its unique identifier is generated; the availability of the main wireless link in the redundant communication channel is detected, and the main wireless link or the standby wired channel is adaptively selected according to the detection result; the alarm data packet is uploaded to the dispatching center through the selected channel.

[0015] The present disclosure, the power quality alarm information is generated according to the transmitted data, and the grid operation decision control is carried out according to the information, further comprising: calculating an event risk control value according to the alarm data packet to determine the priority of the abnormal event; according to the priority, matching the candidate control scheme in the preset knowledge base to constitute the action space; collecting the real-time operation parameters of the distribution network to constitute the state space; constructing a reinforcement learning optimization model according to the action space, the state space and the state transition probability, using the reinforcement learning optimization model to perform policy iteration on the candidate control scheme, stopping iteration when the cumulative reward reaches a preset threshold, and outputting the optimal control scheme; the optimal control scheme is converted into a control instruction and executed to obtain the power quality control result.

[0016] In a second aspect, the present application also provides an electronic device, which comprises: at least one processor; and, a memory in communication with the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the power quality control method of the AC / DC distribution network.

[0017] The power quality management and control method and device of the AC-DC distribution network provided by the application can ensure the accuracy and robustness of feature extraction, and has good comparability and standardization effect. The alarm data packet is uploaded through the redundant communication channel, and the intelligent selection mechanism of the redundant communication channel is combined to ensure that the alarm data can still be reliably transmitted in the case of main link failure or interference, thereby improving the accuracy and response speed of the power quality abnormality alarm, and enhancing the reliability, fault tolerance and security of data transmission. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the application. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0019] Figure 1 A flowchart of a power quality management and control method of an AC-DC distribution network provided by an embodiment of the application is shown. Figure 2 A module diagram of a power quality management and control system of an AC-DC distribution network provided by an embodiment of the application is shown. DETAILED DESCRIPTION

[0020] In order to make those skilled in the art better understand the technical solutions of the present disclosure, and to fully understand and implement the implementation process of the present disclosure how to apply technical means to solve technical problems and achieve corresponding technical effects, the technical solutions in the embodiments of the present disclosure will be described clearly and completely in the embodiments of the present disclosure with reference to the drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all. The embodiments of the present disclosure and each feature in the embodiments can be combined without conflict, and the formed technical solutions are within the protection scope of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative labor should be within the protection scope of the present disclosure.

[0021] It should be noted that the terms "first", "second", and the like in the description and claims of the present disclosure and the foregoing drawings are used to distinguish between similar objects, and are not necessarily used to describe a particular sequential or chronological order. It should be understood that the data thus used can be interchanged, where appropriate, so that the embodiments of the present disclosure described herein can be carried out in sequences other than those illustrated or described herein. In addition, the terms "comprise" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product, or apparatus that includes a list of steps or units as processes, methods, systems, products, or apparatuses are not necessarily limited to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products, or apparatuses.

[0022] The embodiment of the present application provides a kind of AC-DC distribution network power quality management and control method, the execution subject of the AC-DC distribution network power quality management and control method includes but is not limited to at least one of the electronic device of server, terminal and the like can be configured to execute the system provided by the present application.The AC-DC distribution network power quality management and control method can be executed by the software or hardware installed in terminal equipment or server equipment in other words, the server includes but is not limited to: single server, server cluster, cloud server or cloud server cluster etc.The server can be independent server, can also be cloud server that provides cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, content distribution network (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platform.

[0023] Referring to Figure 1 As shown in the figure, a flowchart of the AC-DC distribution network power quality management and control method provided by an embodiment of the present application.In the embodiment, the AC-DC distribution network power quality management and control method includes: S1, collecting power quality data in AC-DC distribution network, and performing feature analysis on power quality data to obtain current feature parameters and voltage abnormality identification; Specifically, this step is: S11: collecting power quality data of a plurality of Internet of Things power quality sensors in AC-DC distribution network, performing harmonic analysis on current waveform of power quality data to obtain current harmonic feature set; S12: performing adaptive threshold judgment on voltage fluctuation of the power quality data, and filtering out voltage abnormal waveform according to the judgment result to generate voltage abnormal label.

[0024] In the embodiments of the present application, the AC-DC power distribution network refers to a power distribution network that simultaneously exists AC (alternating current) and DC (direct current) power supply, for example: in an industrial park: AC side: industrial load, distributed photovoltaic, flexible AC interconnection device; DC side: data center, charging pile, super capacitor energy storage. The Internet of Things power quality sensor is distributed in the AC bus, DC bus, distributed power access point or key load node in the AC-DC power distribution network, and is used to collect power quality data. The power quality data is an index for describing the power supply quality of the power system, including but not limited to: voltage waveform (stability, fluctuation, sudden drop / sudden rise), current waveform (harmonic content, impulse current), power factor, frequency deviation.

[0025] In the AC-DC hybrid power distribution network, the Internet of Things power quality sensor is installed on the AC bus, DC bus, distributed power access point and key load end, and is electrically isolated and calibrated, so that the Internet of Things power quality sensor has node ID and position information, the sampling rate and range are configured according to the running scene, the current and voltage waveforms are sampled at 10 kHz in the normal state, and when disturbance is encountered, it can be adaptively increased to a higher sampling rate, and the anti-aliasing filter is used to ensure data accuracy. The collected power quality data has a timestamp and node information, and a certain length of waveform is retained in the local circular buffer to support backtracking, thereby realizing high-precision and continuous collection of power quality of the power distribution network.

[0026] In detail, the current waveform of the power quality data is analyzed to obtain a current harmonic feature set, including: The power quality data is time-stamped to obtain aligned quality data; The aligned quality data is removed by using a preset adaptive filter to obtain denoised quality data; The denoised quality data is normalized to obtain normalized quality data; The current waveform of the normalized quality data is extracted; Obtain the waveform division length, the window function and the overlap rate, and divide the current waveform into a plurality of initial frame signals according to the waveform division length; The initial frame signals are optimized by using the window function and the overlap rate to obtain analysis frame signals; The analysis frame signals are subjected to fast Fourier transform to obtain the amplitude spectrum and the phase spectrum of each frequency; The fundamental frequency is determined from the amplitude spectrum and the phase spectrum, and the fundamental amplitude and the fundamental phase of the fundamental frequency are identified; Obtain integer multiple frequency points of the fundamental frequency, and extract the multiple frequency point amplitude and the multiple frequency point phase of the integer multiple frequency points; According to the fundamental amplitude and the multiple frequency point amplitude, the harmonic content rate of each integer multiple frequency point is calculated. According to the harmonic content rate, a total harmonic distortion rate is calculated; The fundamental amplitude, the fundamental phase, the frequency multiplication point amplitude, the frequency multiplication point phase, the harmonic content rate and the total harmonic distortion rate form a current harmonic feature set.

[0027] In detail, due to the possible deviation of the sampling time of the Internet of Things power quality sensor, the original power quality data has time dislocation, and the time tags of the power quality data collected by each Internet of Things power quality sensor are interpolated or compensated and corrected with a unified clock source as a reference, so that the data at the same sampling time can correspond on the same time axis, thereby obtaining aligned quality data with consistent timing.

[0028] In the aligned quality data, power frequency interference, electromagnetic environmental noise and sensor measurement error are often mixed, and an adaptive filtering algorithm is used to adjust the filter weight in real time, which can automatically optimize the filtering parameters following the statistical characteristics of the input signal, so as to suppress random noise while maximizing the useful signal characteristics, and finally obtain denoising quality data.

[0029] According to the overall distribution of the denoising quality data, the maximum and minimum values, the mean and the standard deviation of the denoising quality data are obtained, and the denoising quality data is scaled or standardized by using minimum-maximum normalization and Z-score standardization, so that the numerical range of the denoising quality data is mapped to a unified interval or distribution, thereby obtaining normalized quality data.

[0030] By setting the sampling frequency and the time sequence index, the continuous power quality data is restored to the current signal sequence in time sequence, and the curve of the corresponding current amplitude changing with time is extracted in the current signal sequence, thereby obtaining the current waveform.

[0031] According to the waveform division length, the continuous current waveform is sequentially divided into several initial frame signals, and the edges of each initial frame signal are weighted processed by using the set window function to weaken the spectral leakage caused by the boundary effect, wherein the window function can be: rectangular window, Hanning window, Hamming window, Blackman window, etc. According to the overlap rate, the adjacent frames are partially overlapped and spliced, so as to improve the frequency domain analysis accuracy while ensuring the time continuity, and finally obtain the smooth analysis frame signal which can be used for subsequent feature extraction.

[0032] The analysis frame signal is discretely decomposed in the frequency domain by using the fast Fourier transform algorithm, the complex spectrum corresponding to each discrete frequency point is calculated, and the modulus of the spectrum is extracted as the amplitude spectrum and the phase angle as the phase spectrum, so as to realize the comprehensive characterization of the frequency domain features of the current signal.

[0033] After the amplitude spectrum and the phase spectrum are acquired, a frequency component closest to a preset power frequency is searched and determined as a fundamental frequency in a frequency domain range, an amplitude information at a spectrum position corresponding to the fundamental frequency is extracted as a fundamental amplitude, and a phase information is extracted as a fundamental phase. After the fundamental frequency is determined, frequency point positions of integer multiples of the fundamental frequency are sequentially calculated according to an integer multiple relationship of the fundamental frequency, and corresponding frequency component information is extracted in the amplitude spectrum and the phase spectrum, so that a multiple frequency point amplitude and a multiple frequency point phase of each multiple frequency point are obtained.

[0034] The multiple frequency point amplitude and the fundamental amplitude are multiplied to calculate a corresponding harmonic content rate, and a calculation formula is as follows:

[0035] Among them, the harmonic content rate of the integer multiple frequency point, the multiple frequency point amplitude of the integer multiple frequency point, the fundamental amplitude. The harmonic content rate is a key indicator for measuring power quality, and is used to represent the proportion of harmonic energy at a multiple frequency point relative to fundamental energy. The total harmonic distortion rate calculation formula is as follows:

[0036]

[0037] Among them, the harmonic content rate of the integer multiple frequency point, the total number of integer multiple frequency points, the total harmonic distortion rate. The total harmonic distortion rate can reflect the comprehensive level of harmonic energy in the entire current waveform. Through the complete time domain preprocessing and frequency domain analysis process, the fundamental wave and each harmonic feature of the current signal can be accurately extracted on the basis of eliminating noise and normalizing contrast differences, and the power quality condition is quantified in the form of harmonic content rate and total harmonic distortion rate, which ensures the accuracy and robustness of feature extraction, so that the obtained current harmonic feature set not only comprehensively covers the key parameters such as amplitude and phase, but also has good comparability and standardization effect, thereby providing reliable basis for subsequent power quality monitoring, abnormal diagnosis and optimization control.

[0038] For S12: adaptive threshold judgment is performed on the voltage fluctuation of the power quality data, and a voltage abnormal waveform is screened out according to a judgment result to generate a voltage abnormal label, and specifically:

[0039] ​​​In the embodiment of the present application, voltage fluctuation detection is performed on the collected power quality data, the voltage fluctuation limit value under different working conditions is dynamically calculated through an adaptive threshold algorithm, the actual voltage waveform is compared with the voltage fluctuation limit value, if it exceeds the voltage fluctuation limit value range, it is determined as an abnormal waveform, and the corresponding voltage abnormality label is automatically generated, thereby realizing accurate identification and labeling of voltage abnormalities.

[0040] In detail, the adaptive threshold determination of the voltage fluctuation of the power quality data is performed, and the voltage abnormal waveform is screened out according to the determination result, and the voltage abnormality label is generated, comprising: The historical voltage waveform of the historical power quality data is obtained, and the historical voltage mean value and the historical voltage standard deviation are calculated according to the historical voltage waveform; The voltage waveform of the normalized quality data is extracted, and the instantaneous voltage value of each frame of the voltage waveform is obtained; The fluctuation amplitude of each frame of the voltage waveform is calculated according to the historical voltage mean value and the instantaneous voltage value; The dynamic fluctuation threshold range is generated by using the historical voltage mean value, the historical voltage standard deviation and a preset fluctuation factor; It is judged whether the fluctuation amplitude is greater than or equal to the lower limit of the dynamic fluctuation threshold range and less than or equal to the upper limit of the dynamic fluctuation threshold range; If the fluctuation amplitude is greater than or equal to the lower limit of the dynamic fluctuation threshold range and less than or equal to the upper limit of the dynamic fluctuation threshold range, the voltage waveform corresponding to the fluctuation amplitude is determined as a voltage normal waveform; If the fluctuation amplitude is less than the lower limit of the dynamic fluctuation threshold range or greater than the upper limit of the dynamic fluctuation threshold range, the voltage waveform corresponding to the fluctuation amplitude is determined as a voltage abnormal waveform, and a voltage abnormality label is generated according to the fluctuation amplitude.

[0041] In detail, the historical voltage waveform sequence is extracted from the historical power quality data, and the historical voltage waveform sequence is preprocessed to remove abnormal values and noise, and the preprocessed historical voltage waveform is statistically analyzed in the time dimension to calculate the mean value and the standard deviation of the historical voltage waveform sequence, thereby obtaining reference indexes reflecting the historical voltage level and fluctuation characteristics.

[0042] The voltage waveform sequence is extracted from the normalized power quality data, and the voltage waveform sequence is divided into a plurality of analysis frames, and the voltage waveform is processed point by point in each analysis frame to obtain the instantaneous voltage value of the corresponding time point.

[0043] The instantaneous voltage value in each analysis frame is calculated point by point based on the historical voltage mean as a reference benchmark, and the deviation is taken as the fluctuation amplitude of the voltage waveform of the corresponding analysis frame, so as to quantify the fluctuation of each frame of voltage waveform relative to the historical voltage mean.

[0044] The formula for calculating the dynamic fluctuation threshold range by using the historical voltage mean, the historical voltage standard deviation and the preset fluctuation factor is as follows:

[0045]

[0046] wherein, represents the upper limit of the dynamic fluctuation threshold range, represents the fluctuation factor, represents the historical voltage standard deviation, represents the lower limit of the dynamic fluctuation threshold range.

[0047] The fluctuation amplitude of each frame of voltage waveform is compared with the upper and lower limits of the dynamic fluctuation threshold range; when the fluctuation amplitude is within the dynamic fluctuation threshold range, the corresponding voltage waveform is determined to be normal; when the fluctuation amplitude is lower than the lower limit of the dynamic fluctuation threshold range or higher than the upper limit of the dynamic fluctuation threshold range, the corresponding voltage waveform is determined to be abnormal, and the corresponding voltage abnormality label is generated according to the fluctuation amplitude, which can be used to identify the position and degree of the abnormal waveform.

[0048] By combining the historical voltage statistical characteristics with the real-time instantaneous voltage fluctuation, adaptive determination of voltage abnormalities is realized, which does not depend on fixed thresholds, so that the thresholds can be dynamically adjusted according to the grid operation state, thereby improving the accuracy and sensitivity of voltage abnormality detection, and the generated voltage abnormality label can accurately identify the position and degree of the abnormal waveform, providing a reliable basis for power quality analysis, fault location and subsequent regulation, and significantly enhancing the intelligence and reliability of power quality monitoring.

[0049] S3, based on the current characteristic parameters and the voltage abnormality identification, the power quality data is differentially compressed, specifically, the power quality data is classified and compressed according to the current harmonic characteristic set and the voltage abnormality label, and a compressed waveform is obtained.

[0050] In the embodiment of the application, the current harmonic characteristic set of the power quality data and the voltage abnormality label are taken as the classification basis, the power quality data is divided into normal frames and abnormal frames, and appropriate compression algorithm is used to generate a compressed waveform according to the classification result of each type of frame.

[0051] In detail, the power quality data is classified and compressed according to the current harmonic characteristic set and the voltage abnormality label to obtain a compressed waveform, including: When any amplitude of the current harmonic feature set is greater than the preset amplitude threshold, the analysis frame signal corresponding to the amplitude greater than the preset amplitude threshold is marked as a harmonic anomaly; A frame of the power quality data is selected as a target frame; It is judged whether the target frame has the voltage anomaly label or the harmonic anomaly; If the target frame does not have the voltage anomaly label and the harmonic anomaly, the target frame is marked as a normal frame; The power quality data of all the normal frames are differentially encoded to obtain encoded quality data; The encoded quality data is quickly compressed to obtain a compressed normal waveform; If the target frame has the voltage anomaly label or the harmonic anomaly, the target frame is marked as an abnormal frame; The power quality data of all the abnormal frames are losslessly compressed to obtain a compressed abnormal waveform The compressed normal waveform and the compressed abnormal waveform are summarized as a compressed waveform.

[0052] In detail, by judging whether the target frame has the voltage anomaly label or the harmonic anomaly label, frames in which voltage and current anomalies appear simultaneously or individually can be quickly identified, and accurate positioning of power quality anomaly events can be achieved.

[0053] If the target frame does not have the voltage anomaly label and the harmonic anomaly, the target frame is a normal frame. For the frames marked as normal in the power quality data, differential calculation is sequentially performed on each frame of power quality data and the previous frame of power quality data, and the difference value is taken as the encoding result, so as to obtain encoded quality data, effectively reducing data redundancy and storage capacity, while maintaining the change characteristics of the power quality data.

[0054] The encoded quality data is divided into blocks according to a fixed length, and redundancy detection is performed on each encoded quality data block to identify low-amplitude differences or repeated patterns between consecutive frames. Efficient encoding methods such as zero-value compression, entropy encoding or arithmetic encoding are used to compress the low-amplitude differences or repeated patterns, map high-frequency data of the encoded quality data block to short code words, and at the same time retain key power quality characteristics. The compressed blocks are combined to generate a compressed normal waveform.

[0055] If the target frame has the voltage anomaly label or the harmonic anomaly, the target frame is an abnormal frame.

[0056] In detail, the lossless compression of the power quality data of all the abnormal frames to obtain a compressed abnormal waveform comprises: Wavelet transform is performed on the power quality data of all the abnormal frames to obtain a multi-scale wavelet coefficient set; calculate an energy distribution of the set of wavelet coefficients, and obtain an energy value of each wavelet coefficient according to the energy distribution; sort the energy values in descending order, and screen a target set of wavelet coefficients corresponding to energy values greater than a preset accumulated energy proportion in the sorting result; quantize the target set of wavelet coefficients to obtain a set of quantized coefficients; perform entropy coding on the set of quantized coefficients to obtain a set of coded coefficients; perform inverse wavelet transform on the set of coded coefficients to obtain a sparse approximate waveform; obtain an original waveform of power quality data of all the abnormal frames, and subtract the original waveform from the sparse approximate waveform to obtain a residual signal; perform lossless compression on the residual signal to obtain a compressed residual; combine the set of coded coefficients and the compressed residual to obtain a compressed abnormal waveform.

[0057] In detail, wavelet transform is performed on the voltage and current waveforms of each abnormal frame to decompose the original time-domain signal into frequency band components of different scales, thereby obtaining a set of wavelet coefficients of multiple scales, reflecting the local time-domain and frequency characteristics of the signal.

[0058] For the set of wavelet coefficients of each abnormal frame, the energy distribution of the coefficients on each scale and each frequency band is calculated, that is, the square of each wavelet coefficient is taken and accumulated to obtain the energy value of each wavelet coefficient, thereby quantizing the energy distribution characteristics of the signal at different scales. By obtaining the energy value of each wavelet coefficient, the local strength of the abnormal waveform in the time-frequency domain can be reflected.

[0059] The continuous amplitudes in the target set of wavelet coefficients are mapped to a limited number of discrete values according to a preset quantization rule, thereby obtaining a set of quantized coefficients. Quantization not only reduces the precision redundancy of data representation, but also retains the main features of the abnormal waveform.

[0060] An entropy coding method (such as Huffman coding or arithmetic coding) is used to efficiently encode the set of quantized coefficients. High-frequency coefficients are mapped to short code words, and low-frequency coefficients are mapped to long code words, thereby generating a set of coded coefficients. This significantly reduces data redundancy while preserving the signal characteristics as much as possible, thereby achieving efficient compression of power quality data of abnormal frames.

[0061] The set of coded coefficients is reconstructed in the time domain according to the scale and mother wavelet function used in wavelet decomposition to generate a sparse approximate waveform. This can preserve the main features of the power quality signal of abnormal frames while removing redundant information.

[0062] All the original voltage and current waveforms marked as abnormal frames are obtained from the power quality data, and each original waveform is subtracted from the corresponding sparse approximation waveform point by point to obtain a residual signal, which reflects the high-frequency details and local abnormal features that the sparse approximation fails to retain. The residual signal is compressed using a lossless compression algorithm (such as LZ77, DEFLATE or entropy encoding, etc.) to minimize data redundancy and storage space while ensuring information integrity. The set of encoding coefficients obtained previously (the core features of the sparse approximation waveform) is combined with the compressed residual signal to form a complete compressed abnormal waveform.

[0063] By classifying and compressing power quality data into normal frames and abnormal frames, targeted data processing is achieved: normal frames use differential encoding and fast compression to reduce redundancy, enabling efficient storage and transmission; abnormal frames use wavelet decomposition to extract multi-scale features, quantization and entropy encoding to generate sparse approximation waveforms, combined with lossless compression of residual signals, which not only retains the main features and minor details of abnormal waveforms, but also significantly reduces data volume. The compressed normal waveforms and compressed abnormal waveforms are combined into complete compressed waveforms. This classification compression strategy can significantly improve storage and transmission efficiency while ensuring the integrity of key power quality features, providing an efficient and reliable data foundation for power quality analysis, anomaly detection and subsequent control.

[0064] S4, transmit the compressed data and feature information through a redundant channel containing a main and backup communication link, specifically, establish a redundant communication channel between the Internet of Things power quality sensor and the edge computing gateway of the preset substation, the redundant communication channel includes: a 5G wireless link and an optical fiber backup channel.

[0065] In the embodiments of the present application, the edge computing gateway: is set in the local processing node of the preset substation, has data storage, fast analysis and uplink communication functions, and is used for preprocessing and forwarding the data collected by the Internet of Things power quality sensor.

[0066] Redundant communication channel: refers to the configuration of two or more independent communication links at the same time, and supports master-slave switching to ensure the continuity and reliability of communication. 5G wireless link: a high-speed low-latency wireless data channel based on cellular mobile network, serving as the default main link. Optical fiber backup channel: a high-bandwidth channel based on wired optical fiber communication technology, taking over data transmission tasks when the main link is abnormal.

[0067] In detail, in the AC-DC power distribution network, to ensure stable transmission of power quality data from the Internet of Things power quality sensor to the edge computing gateway of the preset substation, the system adopts a redundant communication channel design: by default, the sensor uploads the real-time collected power quality data to the edge gateway through the 5G wireless link, realizing low-latency high-speed transmission; at the same time, the optical fiber backup channel is in standby state and continuously monitors the link. When the 5G wireless link is disturbed, delayed too high or interrupted, the system will automatically switch to the optical fiber backup channel to ensure that the power quality data and alarm information can be transmitted uninterruptedly and reliably to the edge computing gateway.

[0068] By establishing a redundant communication channel of 5G wireless link and optical fiber backup channel between the Internet of Things power quality sensor and the edge computing gateway of the substation, multi-path protection of communication is realized. When the main link fails or is disturbed, the backup channel can immediately take over data transmission, ensuring the continuity and reliability of power quality monitoring data, thereby improving the anti-interference ability of the system, reducing the risk of data loss, and enhancing the stability and security of power quality monitoring and real-time control.

[0069] S5、According to the current harmonic feature set, the voltage anomaly label and the compressed waveform, an alarm data packet is generated and uploaded to a preset dispatch center through the redundant communication channel.

[0070] In the embodiment of the application, in the operation of the AC-DC power distribution network, the alarm data packet not only carries the occurrence type and characteristic parameters of the power quality anomaly, but also carries the compressed original waveform as auxiliary verification information. The alarm data packet is transmitted through a redundant communication channel: under normal circumstances, the 5G wireless link is preferentially used to realize low-latency uploading, and when the 5G link fails or is delayed too high, the system automatically switches to the optical fiber backup channel to ensure that the alarm data can be reliably and in real time transmitted to the preset dispatch center, thereby supporting the dispatch center to quickly make power quality control decisions.

[0071] The preset dispatch center: a centralized control platform located in the park or regional power grid, responsible for receiving alarm data, power quality evaluation and decision control, and issuing dispatch instructions to energy storage devices, flexible interconnected devices, etc.

[0072] In detail, the alarm data packet is generated according to the current harmonic feature set, the voltage anomaly label and the compressed waveform, and uploaded to the preset dispatch center through the redundant communication channel, comprising: The current harmonic feature set, the voltage anomaly label and the compressed waveform are dynamically weighted to obtain a fusion data set; The fusion data set is formatted and packaged to obtain a packaged data set; According to the compressed waveform, a compressed fingerprint code is generated, and the compressed fingerprint code is embedded in the encapsulation data set to obtain an updated data set; A unique identifier of the updated data set is acquired, and an alarm data packet is generated according to the unique identifier and the updated data set; The real-time availability of the 5G wireless link in the redundant communication channel is detected to obtain a detection result; When the detection result is that the 5G wireless link is available, the 5G wireless link of the redundant communication channel is selected; When the detection result is that the 5G wireless link is unavailable, the optical fiber backup channel of the redundant communication channel is selected; The alarm data packet is uploaded to a preset dispatching center through the redundant communication channel.

[0073] In detail, the weighted data is fused to generate a fusion data set comprehensively reflecting the power quality state by real-time adjustment of weights according to the current harmonic feature set, the voltage abnormality label, and the data importance, abnormality degree, or time sensitivity of the compressed waveform.

[0074] The fusion data set is processed according to a unified data structure and protocol, necessary identification fields, time stamps, frame sequence numbers, and check codes, and other metadata are added to various types of information of the fusion data set, and the current harmonic features, the voltage abnormality label, and the compressed waveform are sequentially integrated in standardized data units, so as to generate an encapsulation data set.

[0075] The compressed waveform is feature-extracted, and indexes capable of representing key features of the waveform, such as main frequency components, amplitude characteristics, or compression coefficient modes, are selected, and a unique compressed fingerprint code is generated based on these features to identify the core information and integrity of the compressed waveform. The generated compressed fingerprint code is embedded in a specified field of the encapsulation data set, and the metadata (such as the time stamp, the frame sequence number, or the check information) is updated to form an updated data set.

[0076] A unique identifier is extracted from the updated data set, the unique identifier can be used to uniquely identify the updated data set and the corresponding power quality state, based on the unique identifier and the content of the updated data set, the key information (such as the current harmonic feature, the voltage abnormality label, the compressed waveform, and the compressed fingerprint code) is integrated and an alarm data packet is generated according to an alarm protocol.

[0077] The 5G wireless link in the redundant communication channel is monitored in real time. By continuously collecting performance indicators such as link signal strength, delay, and packet loss rate, the current availability of the 5G wireless link is determined and the detection results are generated: when the detection results show that the 5G wireless link is available, the 5G wireless link is used as the main transmission channel to send alarm data packets to ensure low-latency and high-bandwidth real-time transmission; if the detection results show that the 5G link is unavailable or the performance drops to an unacceptable threshold, it will automatically switch to the optical fiber backup channel, using the stability and high reliability of the optical fiber to ensure uninterrupted data transmission, and upload the alarm data packet to the preset dispatching center through the selected channel. At the same time, the transmission path and status information are recorded to ensure that the power quality alarm information can be continuously and reliably transmitted, thereby achieving high availability and fault tolerance of the monitoring system.

[0078] By dynamically weighting and formatting current harmonic characteristics, voltage anomaly labels, and compressed waveforms, comprehensive expression and structured management of power quality data are achieved. By generating compressed fingerprint codes and embedding them into encapsulated data sets, and using unique identifiers to generate alarm data packets, rapid unique identification and integrity verification of abnormal events are achieved. Combined with the intelligent selection mechanism of redundant communication channels (5G main link and optical fiber backup channel), it ensures that alarm data can still be reliably transmitted to the dispatching center in the event of main link failure or interference. This not only improves the accuracy and response speed of power quality anomaly alarms, but also enhances the reliability, fault tolerance, and security of data transmission, providing efficient and robust technical support for real-time monitoring and intelligent control of power grids.

[0079] S6. Utilize the preset dispatching center that receives the alarm data packet to make decisions and control the power quality data to obtain a power quality target control result.

[0080] In an embodiment of the present invention, after receiving the alarm data packet from the sensor end, the preset dispatching center parses the current harmonic characteristics, voltage anomaly labels and compressed waveform information therein, and combines historical power quality data and real-time operating parameters to conduct a comprehensive assessment of the current state of the power grid. Based on the preset control strategy and reinforcement learning optimization model, a decision analysis is performed on possible harmonics, fluctuations and abnormal events, and a targeted adjustment plan is generated, thereby obtaining the power quality target control result.

[0081] In detail, the preset dispatching center that receives the alarm data packet performs decision-making control on the power quality data to obtain the power quality target control result, including: generating an event risk control value of the power quality data according to the alarm data packet; determining the abnormal event priority of the power quality data according to the event risk control value; According to the event priority, a corresponding candidate control scheme is matched in a preset double-layer knowledge base, and the candidate control scheme is taken as an action space; A parameter operating state of the AC-DC distribution network is collected, and the parameter operating state is taken as a state space; One candidate control scheme in the action space is selected as a target action, and one parameter operating state in the state space is selected as a target state, and the target action and the target state are taken as a target analysis group; A state transition probability of the target state in the target analysis group and an immediate reward of the target analysis group are obtained; According to the action space, the state space, the state transition probability, the immediate reward, and a preset discount factor, a reinforcement learning optimization model is constructed; The reinforcement learning optimization model is used for policy iteration of the candidate control scheme, the number of iterations is counted, and a final cumulative reward is calculated according to the immediate reward and the number of iterations; When the final cumulative reward reaches a preset reward threshold, the iteration is stopped, and a final candidate control scheme is taken as an optimal control scheme; The optimal control scheme is converted into a control instruction, and the control instruction is executed on the power quality data to obtain an executed power quality target control result.

[0082] In detail, after receiving the alarm data packet, the dispatching center parses the key information such as the current harmonic feature set, the voltage abnormality label, and the compressed waveform in the alarm data packet, and combines the time stamp of the abnormal event, the abnormal amplitude, and the historical statistical law to quantitatively evaluate the risks such as power grid fluctuation, equipment loss, or power supply interruption that may be caused by the abnormal event. The corresponding event risk control value is calculated by a risk score based on weight or a probability statistical model.

[0083] The event risk control value is compared with a preset risk threshold interval, and the abnormal event in the power quality data is prioritized: if the event risk control value is less than the lower limit of the risk threshold interval, the event corresponding to the event risk control value is marked as low priority and only recorded or delayed; if the event risk control value is within the risk threshold interval, the event corresponding to the event risk control value is marked as medium priority and enters the conventional dispatching and control process; if the event risk control value is greater than the upper limit of the risk threshold interval, the event corresponding to the event risk control value is marked as high priority, triggers an emergency alarm, and preferentially allocates governance and dispatching resources.

[0084] The system uses the priority of abnormal events as the retrieval condition and enters the preset two-layer knowledge base for matching. The first-layer knowledge base stores general control strategies based on rules and expert experience, and the second-layer knowledge base contains personalized control solutions generated by combining historical data, mechanism models and intelligent optimization algorithms. The system first quickly locates the initial candidate solutions that meet the event type and priority requirements in the first-layer knowledge base, and then further screens out more targeted and optimal candidate control solutions in the second-layer knowledge base.

[0085] All candidate control solutions are recorded as action space , the parameter operation state of the AC / DC distribution network is recorded as the state space , combine candidate control schemes and parameter operating states one by one, and combine as the target analysis group.

[0086] Using the state transition probability of the target state in the target analysis group , instant rewards , discount factor Constructing a Markov Decision Process , a reinforcement learning optimization model is constructed through Markov decision process and strategy iteration.

[0087] Policy iteration is performed on the reinforcement learning optimization model to continuously update the decision strategy of the candidate control scheme. The calculation formula is as follows:

[0088] in, Indicates that the parameter is running. Candidate control scheme The number of sampling times, Indicates the The instant reward obtained by subsampling, Indicates that the parameter is running. Candidate control schemes Instant rewards.

[0089]

[0090] in, Indicates that the parameter is running. Candidate control schemes The number of sampling times, Indicates that the parameter is running. Candidate control scheme And transfer to parameter running state The number of times, Indicates that the parameter is running. Candidate control schemes And transfer to parameter running state The state transition probability.

[0091]

[0092] in, Indicates that the parameter is running. Candidate control schemes strategy, Indicates immediate reward, represents the discount factor, Indicates that the parameter is running. Candidate control schemes And transfer to parameter running state The state transition probability, Indicates that the parameter is running. Next, according to the strategy The expected cumulative reward of the action, Indicates that the parameter is running. Next, according to the strategy The expected cumulative reward for the action.

[0093]

[0094] in, Indicates that the parameter is running. Candidate control schemes Instant rewards, represents the discount factor, Indicates that the parameter is running. Candidate control schemes And transfer to parameter running state The state transition probability, Indicates that the parameter is running. Next, according to the strategy The expected cumulative reward of the action, Indicates that the parameter is running. Candidate control schemes The expected cumulative reward.

[0095]

[0096] in, Indicates that the parameter is running. The optimal control solution selected is Indicates that the parameter is running. Candidate control schemes The expected cumulative reward of Indicates selection The largest candidate control solution .

[0097] The number of iterations is counted. During the iteration process, the final cumulative reward is calculated based on the immediate rewards obtained in each round and the corresponding number of iterations. When the final cumulative reward reaches the reward threshold, the strategy is considered to have converged, the iteration is stopped, and the candidate control scheme obtained at this time is determined as the optimal control scheme. The calculation formula is as follows:

[0098] in, represents the number of iterations, Indicates the The immediate reward at the time step, Indicates the The discount factor for the time step, From the first iteration to the The final cumulative reward for iterations.

[0099] The obtained optimal control scheme is converted into control instructions that can be directly issued, acting on the power quality data of the distribution network, and the power quality data is adjusted and optimized in real time, so as to obtain the power quality target management and control results after execution, and realize effective improvement and optimization control of power quality.

[0100] Through the hierarchical management of event risk control values ​​and abnormal event priorities, the scientificity and rationality of alarm responses can be ensured; a two-layer knowledge base is used to match candidate control schemes, and a reinforcement learning optimization model is combined to perform strategy iteration in state-action interaction, so that the control strategy can be continuously optimized and gradually converged to the optimal in a dynamic environment, thereby improving the accuracy and adaptability of the control scheme; the optimal control scheme is converted into control instructions and directly acts on power quality data, realizing closed-loop management and control from risk perception, strategy optimization to execution feedback, which not only improves the stability and reliability of power quality, but also enhances the intelligent management and control capabilities of the distribution network.

[0101] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0102] like Figure 2 FIG. 1 is a functional module diagram of an AC / DC power distribution network power quality control system provided by an embodiment of the present invention.

[0103] In an embodiment of the present disclosure, a system for controlling the power quality of an AC / DC distribution network is provided. The system for controlling the power quality of an AC / DC distribution network corresponds to the method for controlling the power quality of an AC / DC distribution network in the above embodiment. Figure 2As shown, the power quality management and control system 100 of the alternating current-direct current power distribution network can be installed in an electronic device. According to the functions to be realized, the power quality management and control system 100 of the alternating current-direct current power distribution network includes a current harmonic analysis module 101, a voltage abnormality determination module 102, a data classification and compression module 103, a redundant communication channel establishment module 104, an alarm data packet generation module 105, and an alarm data management and control module 106. The detailed descriptions of the functional modules are as follows: The current harmonic analysis module 101 is configured to collect power quality data of a plurality of Internet of Things power quality sensors in the alternating current-direct current power distribution network, perform harmonic analysis on current waveforms of the power quality data, and obtain a current harmonic feature set. The voltage abnormality determination module 102 is configured to perform adaptive threshold determination on voltage fluctuations of the power quality data, and filter out voltage abnormal waveforms according to the determination results to generate voltage abnormality labels. The data classification and compression module 103 is configured to classify and compress the power quality data according to the current harmonic feature set and the voltage abnormality labels to obtain compressed waveforms. The redundant communication channel establishment module 104 is configured to establish a redundant communication channel between the Internet of Things power quality sensors and an edge computing gateway of a preset substation, wherein the redundant communication channel includes a 5G wireless link and an optical fiber backup channel. The alarm data packet generation module 105 is configured to generate an alarm data packet according to the current harmonic feature set, the voltage abnormality labels, and the compressed waveforms, and upload the alarm data packet to a preset dispatch center through the redundant communication channel. The alarm data management and control module 106 is configured to perform decision control on the power quality data by using the preset dispatch center receiving the alarm data packet to obtain power quality target management and control results.

[0104] In an embodiment, the current harmonic analysis module 101 performs harmonic analysis on current waveforms of the power quality data to obtain a current harmonic feature set, including: aligning time stamps of the power quality data to obtain aligned quality data; removing noise from the aligned quality data by using a preset adaptive filter to obtain denoised quality data; normalizing the denoised quality data to obtain normalized quality data; extracting current waveforms of the normalized quality data; obtaining waveform division length, window function, and overlap rate, and dividing the current waveforms into a plurality of initial frame signals according to the waveform division length; optimizing the initial frame signals by using the window function and the overlap rate to obtain analysis frame signals. performing fast Fourier transform on the analysis frame signal to obtain an amplitude spectrum and a phase spectrum of each frequency; determining a fundamental frequency from the amplitude spectrum and the phase spectrum, and identifying a fundamental amplitude and a fundamental phase of the fundamental frequency; obtaining integer multiple frequency points of the fundamental frequency, and extracting a multiple amplitude and a multiple phase of the integer multiple frequency points; calculating a harmonic content rate of each of the integer multiple frequency points according to the fundamental amplitude and the multiple amplitude; calculating a total harmonic distortion rate according to the harmonic content rate; forming a current harmonic feature set including the fundamental amplitude, the fundamental phase, the multiple amplitude, the multiple phase, the harmonic content rate, and the total harmonic distortion rate.

[0105] In an embodiment, the voltage anomaly determination module 102 performs adaptive threshold determination on voltage fluctuation of the power quality data, and screens out voltage abnormal waveform according to the determination result, generates a voltage anomaly label, including: obtaining a historical voltage waveform of historical power quality data, and calculating a historical voltage mean and a historical voltage standard deviation according to the historical voltage waveform; extracting a voltage waveform of the normalized quality data, and obtaining an instantaneous voltage value of each frame of the voltage waveform; calculating a fluctuation amplitude of each frame of the voltage waveform according to the historical voltage mean and the instantaneous voltage value; generating a dynamic fluctuation threshold range using the historical voltage mean, the historical voltage standard deviation, and a preset fluctuation factor; determining whether the fluctuation amplitude is greater than or equal to a lower limit of the dynamic fluctuation threshold range, and less than or equal to an upper limit of the dynamic fluctuation threshold range; if the fluctuation amplitude is greater than or equal to the lower limit of the dynamic fluctuation threshold range, and less than or equal to the upper limit of the dynamic fluctuation threshold range, the voltage waveform corresponding to the fluctuation amplitude is determined as a voltage normal waveform; if the fluctuation amplitude is less than the lower limit of the dynamic fluctuation threshold range, or greater than the upper limit of the dynamic fluctuation threshold range, the voltage waveform corresponding to the fluctuation amplitude is determined as a voltage abnormal waveform, and a voltage anomaly label is generated according to the fluctuation amplitude.

[0106] In an embodiment, the data classification compression module 103 performs classification compression on the power quality data according to the current harmonic feature set and the voltage anomaly label, and obtains a compressed waveform, including: When any of the amplitude values of the current harmonic feature set is greater than a preset amplitude threshold value, the analysis frame signal corresponding to the amplitude value greater than the preset amplitude threshold value is marked as a harmonic anomaly; A frame of the power quality data is selected as a target frame; It is judged whether the target frame has the voltage anomaly label or the harmonic anomaly; If the target frame does not have the voltage anomaly label and the harmonic anomaly, the target frame is marked as a normal frame; The power quality data of all the normal frames is differentially encoded to obtain encoded quality data; The encoded quality data is quickly compressed to obtain a compressed normal waveform; If the target frame has the voltage anomaly label or the harmonic anomaly, the target frame is marked as an abnormal frame; The power quality data of all the abnormal frames is losslessly compressed to obtain a compressed abnormal waveform The compressed normal waveform and the compressed abnormal waveform are summarized as a compressed waveform.

[0107] In an embodiment, the data classification compression module 103 performs classification compression on the power quality data according to the current harmonic feature set and the voltage anomaly label to obtain a compressed waveform, including: The power quality data of all the abnormal frames is wavelet transformed to obtain a multi-scale wavelet coefficient set; The energy distribution of the wavelet coefficient set is calculated, and the energy value of each wavelet coefficient is obtained according to the energy distribution; The energy values are sorted in descending order, and a target wavelet coefficient set corresponding to the energy values greater than a preset cumulative energy proportion is selected from the sorting result; The target wavelet coefficient set is quantized to obtain a quantized coefficient set; The quantized coefficient set is entropy encoded to obtain an encoded coefficient set; The encoded coefficient set is inverse wavelet transformed to obtain a sparse approximate waveform; The original waveform of the power quality data of all the abnormal frames is obtained, and the original waveform is subtracted from the sparse approximate waveform to obtain a residual signal; The residual signal is losslessly compressed to obtain a compressed residual; The encoded coefficient set and the compressed residual are combined as a compressed abnormal waveform.

[0108] In an embodiment, the alarm data packet generation module 105 generates an alarm data packet according to the current harmonic feature set, the voltage anomaly label, and the compressed waveform, and uploads the alarm data packet to a preset dispatch center through the redundant communication channel, including: dynamically weighting the current harmonic feature set, the voltage anomaly label, and the compressed waveform to obtain a fusion data set; formatting and packaging the fusion data set to obtain a packaged data set; generating a compressed fingerprint code according to the compressed waveform, and embedding the compressed fingerprint code in the packaged data set to obtain an updated data set; obtaining a unique identifier of the updated data set, and generating an alarm data packet according to the unique identifier and the updated data set; detecting the real-time availability of the 5G wireless link in the redundant communication channel to obtain a detection result; when the detection result is that the 5G wireless link is available, selecting the 5G wireless link of the redundant communication channel; when the detection result is that the 5G wireless link is not available, selecting the optical fiber backup channel of the redundant communication channel; uploading the alarm data packet to a preset dispatch center through the redundant communication channel.

[0109] In an embodiment, the alarm data control module 106 performs decision control on the power quality data using a preset dispatch center that receives the alarm data packet to obtain power quality target control results, including: generating an event risk control value of the power quality data according to the alarm data packet; determining an abnormal event priority of the power quality data according to the event risk control value; matching a corresponding candidate control scheme in a preset double-layer knowledge base according to the event priority, and taking the candidate control scheme as an action space; collecting parameter operating states of an AC-DC distribution network, and taking the parameter operating states as a state space; selecting one candidate control scheme in the action space as a target action, and one parameter operating state in the state space as a target state, and taking the target action and the target state as a target analysis group; obtaining a state transition probability of the target state in the target analysis group, and an immediate reward of the target analysis group; constructing a reinforcement learning optimization model according to the action space, the state space, the state transition probability, the immediate reward, and a preset discount factor; The reinforcement learning optimization model is used for policy iteration on the candidate control scheme, the number of iterations is counted, and a final cumulative reward is calculated according to the instant reward and the number of iterations; When the final cumulative reward reaches a preset reward threshold, iteration is stopped, and a final candidate control scheme is taken as an optimal control scheme; The optimal control scheme is converted into a control instruction, and the control instruction is executed on the power quality data to obtain an executed power quality target management and control result.

[0110] The voltage fluctuation of the power quality data is adaptively threshold determined, and a voltage abnormal waveform is screened out according to a determination result to generate a voltage abnormal label, without relying on a fixed threshold, so that the threshold can be dynamically adjusted according to the operation state of the power grid, thereby improving the accuracy and sensitivity of voltage abnormality detection; the power quality data is classified and compressed according to the current harmonic feature set and the voltage abnormal label to obtain a compressed waveform; normal frames utilize differential encoding and fast compression to reduce redundancy, achieving efficient storage and transmission; abnormal frames extract multi-scale features through wavelet decomposition, quantization and entropy encoding to generate a sparse approximate waveform, and combined with lossless compression of residual signals, the main features and tiny details of the abnormal waveform are retained, and the data amount is significantly reduced; the compressed normal waveform and the compressed abnormal waveform are summarized as a complete compressed waveform; a redundant communication channel is established between the Internet of Things power quality sensor and the edge computing gateway of the preset substation, the redundant communication channel includes a 5G wireless link and an optical fiber backup channel, an alarm data packet is generated according to the current harmonic feature set, the voltage abnormal label and the compressed waveform, and the alarm data packet is uploaded to a preset dispatch center through the redundant communication channel, combined with an intelligent selection mechanism (5G main link and optical fiber backup channel) of the redundant communication channel, the alarm data can still be reliably transmitted to the dispatch center in the case of main link failure or interference, not only improving the accuracy and response speed of the power quality abnormality alarm, but also enhancing the reliability, fault tolerance and security of data transmission; the preset dispatch center receiving the alarm data packet makes a decision control on the power quality data, combined with a reinforcement learning optimization model for policy iteration in state-action interaction, so that the control strategy can be continuously optimized in a dynamic environment and gradually converges to an optimal value, thereby improving the precision and adaptability of the control scheme, obtaining a power quality target management and control result, improving the monitoring coverage, real-time performance and reliability, and effectively meeting the demand of modern AC-DC hybrid distribution network for high-precision, high-reliability and fast-response power quality management and control.

[0111] The specific limitation of the power quality management and control system of the AC-DC power distribution network can refer to the limitation of the power quality management and control method of the AC-DC power distribution network in the above, and will not be described here. Each module in the power quality management and control system of the AC-DC power distribution network can be realized by software, hardware and their combination. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so that the processor calls and executes the operations corresponding to each module.

[0112] In one embodiment, a computer device is provided, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Collecting power quality data of a plurality of Internet of Things power quality sensors in an AC-DC power distribution network, performing harmonic analysis on the current waveform of the power quality data to obtain a current harmonic feature set; Performing adaptive threshold judgment on the voltage fluctuation of the power quality data, and filtering out voltage abnormal waveforms according to the judgment result to generate voltage abnormal labels; Classifying and compressing the power quality data according to the current harmonic feature set and the voltage abnormal label to obtain a compressed waveform; Establishing a redundant communication channel between the Internet of Things power quality sensor and the edge computing gateway of a preset substation, the redundant communication channel comprising a 5G wireless link and an optical fiber backup channel; Generating an alarm data packet according to the current harmonic feature set, the voltage abnormal label and the compressed waveform, and uploading the alarm data packet to a preset dispatching center through the redundant communication channel; Using the preset dispatching center receiving the alarm data packet to perform decision control on the power quality data to obtain power quality target control results.

[0113] In several embodiments provided by the present application, it should be understood that the disclosed devices and systems can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of the modules is only a logical function division. In actual implementation, there can be another division manner.

[0114] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of hardware plus software functional modules.

[0115] Therefore, embodiments should be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are intended to be embraced therein. No reference is intended to be made to any disclaimer, unless the following language appear in a claim: "except as claimed".

[0116] It will be apparent to those skilled in the art that the application is not limited to the details of the above-exemplified embodiments and that the application can be implemented in other concrete forms without departing from the spirit or essential characteristics thereof.

[0117] In some embodiments of the present embodiment, a computer readable storage medium is provided, and a computer program is stored on the computer readable storage medium, and the computer program is characterized in that the computer program is executed by a processor to implement the steps of the method described in the above embodiment.

[0118] The readable storage medium of the present application stores a computer program, and the computer program can realize the following when executed by a processor of an electronic device: Collecting power quality data of a plurality of Internet of Things power quality sensors in an AC-DC power distribution network, performing harmonic analysis on current waveforms of the power quality data to obtain a current harmonic feature set; Performing adaptive threshold judgment on voltage fluctuations of the power quality data, and filtering out voltage abnormal waveforms according to the judgment result to generate a voltage abnormal label; Classifying and compressing the power quality data according to the current harmonic feature set and the voltage abnormal label to obtain a compressed waveform; Establishing a redundant communication channel between the Internet of Things power quality sensor and an edge computing gateway of a preset substation, the redundant communication channel including a 5G wireless link and an optical fiber backup channel; Generating an alarm data packet according to the current harmonic feature set, the voltage abnormal label and the compressed waveform, and uploading the alarm data packet to a preset dispatch center through the redundant communication channel; Using a preset dispatch center receiving the alarm data packet to perform decision control on the power quality data to obtain a power quality target control result.

[0119] It should be noted that the functions or steps that the computer readable storage medium or the computer device can realize correspond to the descriptions of the server side and the client side in the foregoing method embodiments, and to avoid repetition, they will not be described one by one here.

[0120] The computer-readable storage medium can also store at least one computer executable program / instruction, for example, computer-readable instructions, which are executable by a computer. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM), cache memory, and / or the like. The computer-readable storage medium may, for example, include read-only memory (ROM), a hard disk, a flash memory, and / or the like. For example, the non-transitory computer-readable storage medium can be connected to a computing device such as a computer, and then, in a case where the computing device executes the computer-readable instructions stored on the computer-readable storage medium, each of the methods described above can be performed.

[0121] In addition to this, the computer device can also include, but is not limited to, a data bus, an input / output (I / O) bus, a display, and an input / output device (for example, a keyboard, a mouse, a speaker, and / or the like), and / or the like.

[0122] The processor can communicate with an external device via a wired or wireless network through the I / O bus.

[0123] In one embodiment, the at least one computer executable instruction can also be compiled or constitute a software product / computer program product, wherein one or more computer executable instructions are executed by the processor to perform the steps of each function and / or method in the embodiments described in the present technology.

[0124] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, database or other medium used in each embodiment provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0125] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above-mentioned division of functional units and modules is exemplified, and in actual application, the above-mentioned functions can be completed by different functional units and modules according to needs, that is, the internal structure of the system is divided into different functional units or modules to complete all or part of the above-mentioned functions.

[0126] In the embodiments provided in the present disclosure, it should be understood that the disclosed system and method can also be implemented in other manners. The above described system embodiments are merely illustrative, for example, the flowcharts and block diagrams in the accompanying drawings show possible implementation architectures, functions and operation of the system, method and computer program product according to the embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams can represent a module, a segment or a portion of code which comprises one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions shown in the blocks can occur in a different order than that shown in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, or they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system, or can be implemented by a combination of dedicated hardware and computer instructions.

[0127] It should be noted that in the present disclosure, the terms "comprising", "containing" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that comprises a list of elements not only includes those elements, but also includes other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, the element limited by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising the element.

[0128] The above described embodiments are merely used to illustrate the technical solutions of the present application, rather than limiting them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for controlling power quality of an AC / DC distribution network, characterized in that: The method comprises: Collect power quality data from AC and DC distribution networks, perform feature analysis on the data, and obtain current characteristic parameters and voltage anomaly indicators; Based on current characteristic parameters and voltage anomaly indicators, power quality data is differentially compressed; Transmit compressed data and feature information through redundant channels including primary and backup communication links; Power quality alarm information is generated based on the transmitted data, and grid operation decision-making and control are carried out based on this information.

2. The AC / DC power distribution network power quality control method according to claim 1, characterized in that: The feature analysis includes: Perform harmonic analysis on the current waveform and extract harmonic characteristic parameters; Perform adaptive threshold analysis on voltage fluctuations and generate voltage anomaly indicators.

3. The AC / DC power distribution network power quality control method according to claim 2, characterized in that: Differential compression processing includes: Based on the current characteristic parameters and voltage anomaly indicators, the abnormal status of each data frame in the power quality data is judged; according to the judgment results, a lossy compression strategy is adopted for the normal data frame, and a lossless compression strategy is adopted for the abnormal data frame, and the compressed data are merged and output.

4. The AC / DC power distribution network power quality control method according to claim 2, characterized in that: The current waveform is subjected to harmonic analysis to extract harmonic characteristic parameters, specifically: Preprocess the power quality data to obtain a standardized current waveform signal; Performing frame processing on the standardized current waveform signal to obtain an analysis frame signal; Perform spectrum analysis on the analysis frame signal to obtain the amplitude and phase parameters of the fundamental frequency and its harmonic frequencies; The content rate of each harmonic and the total harmonic distortion rate are calculated based on the amplitude and phase parameters.

5. The AC / DC power distribution network power quality control method according to claim 4, characterized in that: The adaptive threshold analysis of voltage fluctuation is performed to generate a voltage anomaly indicator, specifically: Obtain historical voltage waveforms and calculate historical voltage mean and historical voltage standard deviation; Collect voltage waveform in real time and extract instantaneous voltage value; Calculate the fluctuation amplitude based on the historical voltage average and instantaneous voltage value; Generate dynamic fluctuation threshold range based on historical voltage mean, historical voltage standard deviation and preset fluctuation factor; determining whether the fluctuation amplitude is within the dynamic fluctuation threshold range; If it is within the range, it is determined to be a normal voltage waveform; if it is out of the range, it is determined to be an abnormal voltage waveform and a voltage abnormality label is generated.

6. The AC / DC power distribution network power quality control method according to claim 5, characterized in that: The differential compression processing of the power quality data includes: Determine frame by frame whether a target frame in the power quality data is marked as voltage anomaly or harmonic anomaly, wherein the harmonic anomaly is obtained by comparing the amplitude of the current harmonic feature with a preset amplitude threshold; If the target frame has no voltage anomaly and harmonic anomaly, it is marked as a normal frame, and differential encoding and lossy compression are performed on it to obtain a compressed normal waveform; If there is any abnormality in the target frame, it is marked as an abnormal frame and losslessly compressed to obtain a compressed abnormal waveform; all compressed normal waveforms and compressed abnormal waveforms are summarized to generate a compressed waveform.

7. The AC / DC power distribution network power quality control method according to claim 6, characterized in that: The specific method for obtaining the compressed abnormal waveform is as follows: Perform wavelet transform on abnormal frame power quality data to obtain a multi-scale wavelet coefficient set; Based on the energy distribution of the wavelet coefficient set, a significant coefficient set with significant energy is screened; Quantizing and losslessly encoding the significant coefficient set to obtain a coded coefficient set; Performing an inverse transform according to the coding coefficients to generate a sparse approximate waveform, and calculating a residual signal between the original waveform and the sparse approximate waveform; Performing lossless compression on the residual signal to obtain a compressed residual; The coding coefficients are combined with the compression residual to form a compressed anomaly waveform.

8. The AC / DC power distribution network power quality control method according to claim 7, characterized in that: Generating power quality warning information according to the transmitted data and making grid operation decision-making and control based on the information includes: Perform weighted fusion of current characteristic parameters, voltage anomaly indicators, and compressed waveforms to generate a fused data set; Format and encapsulate the fused data set, and embed the compressed fingerprint code generated based on the compressed waveform to obtain the updated data set, and generate an alarm data packet containing the updated data set and its unique identifier; Detect the availability of the primary wireless link in the redundant communication channel and adaptively select the primary wireless link or backup wired channel based on the detection results; upload the alarm data packet to the dispatch center through the selected channel.

9. The AC / DC power distribution network power quality control method according to claim 1 or 6, characterized in that: The generating of power quality alarm information according to the transmitted data and performing power grid operation decision control based on the information also includes: Calculate event risk control values ​​based on alarm data packets and determine the priority of abnormal events; According to the priority, candidate control schemes are matched in the preset knowledge base to form an action space; Collect real-time operating parameters of the distribution network to form a state space; A reinforcement learning optimization model is constructed based on the action space, state space, and state transition probability. The reinforcement learning optimization model is used to iterate the strategies of the candidate control schemes. When the cumulative reward reaches the preset threshold, the iteration is stopped and the optimal control scheme is output. The optimal control scheme is converted into control instructions and executed to obtain the power quality control results.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for controlling power quality of an AC / DC distribution network according to any one of claims 1 to 9.

Images