CN111781422B

CN111781422B - Synchronous phasor measurer for power distribution network

Info

Publication number: CN111781422B
Application number: CN202010838033.9A
Authority: CN
Inventors: 孙煜皓
Original assignee: Jianke Yunzhi Shenzhen Technology Co ltd
Current assignee: Jianke Yunzhi Shenzhen Technology Co ltd
Priority date: 2020-08-19
Filing date: 2020-08-19
Publication date: 2023-04-18
Anticipated expiration: 2040-08-19
Also published as: CN111781422A

Abstract

A power distribution network synchronous phasor measurer comprises a main processor, a coprocessor, a first memory card, a second memory card, an Ethernet port, a time service module, an analog-to-digital converter and an edge calculation module; and the edge calculation module is connected with the main processor and used for calling phasor data to calculate and returning a calculation result to the main processor. According to the synchronous phasor measurer for the power distribution network, due to the fact that the edge computing module is introduced, dependence on cloud can be eliminated, large data generated by the edge sensor does not need to be transmitted back to the cloud server, rapid operation can be conducted, and the requirement for data transmission is lowered.

Description

Synchronous phasor measurer for power distribution network

Technical Field

The invention relates to a synchronous phasor measurer for a power distribution network.

Background

In recent years, with the use of a large number of industrial and household appliances in China, the power distribution network has the characteristics of large scale and complex structure. The realization of the active power distribution network and the incorporation of the distributed power supply into the power distribution network lead the complexity of the dynamic behavior of the power distribution network structure and the power distribution network operation process to be higher and higher. Therefore, predicting the power utilization trend of the power distribution system in real time, judging the abnormal operation of the power distribution network, and taking effective measures to the detection result to reduce the failure rate of the power distribution network become an indispensable part in the field of power system control.

With the rapid development of scientific technology, the wide use of the global positioning system provides accurate time information for Phasor Measurement, and in order to dynamically monitor and protect a power distribution network in real time in each country, more and more synchronous Phasor Measurement units PMU (Phasor Measurement Unit) play an important role in an actual power grid. The PMU can collect the voltage and the current of the line in real time, and the real-time phase angle of the voltage and the current can be calculated by combining the time information provided by the GPS.

The first PMU in the world was developed in 1993 and entered into practical use. The technology of the global positioning system, which is developed in the 80 th century in the united states, is gradually mature, and can provide high time service precision for a large number of users. The real application of GPS in the power system is from 1993, and can meet the requirements of monitoring, protection and control in the power system and meet the requirement of PMU remote synchronization.

The research of phase angle measurement has been started from the 80 s of the last century, switzerland realizes the time service function by adopting radio signals, and the phase angle measurement adopts a zero crossing point detection method. In 1983, the radio signal is also used as a synchronous clock in the United states, and the method of solving the phase angle of the voltage and the current by using a symmetrical component discrete Fourier transform method is proposed.

In France, a PMU is installed at a high-voltage key node in a national power grid in 1997 to measure a node voltage phasor, the obtained phasor is used for monitoring the running state of the power grid in real time, and relevant treatment measures can be quickly taken when a dangerous accident occurs to part of the power grid so as to reduce the harm caused by the accident.

In addition, many countries such as the united kingdom and japan have made extensive research on PMU development and application.

The development of synchronized phasor measurement techniques in various countries requires corresponding technical specifications to be unified, thereby ensuring that the developed devices are compatible with the power system. IEEE in the United states released a revision of IEEE Std PC37.118-005 in 2005. The standard specifies data formats and clock selections in detail.

More than 1995, the north China electric power started the related research of domestic PMU. 1995 developed a PMU, which measures a phase angle by using a test, GPS time service and zero-crossing point detection method, and the minimum measurement accuracy of the phase angle of the developed device is 0.1 degree after passing through an experimental voltage and current in an actual power grid.

The global network synchronous detection system based on GPS and PMU is proposed by Qinghua university in 1997. The system can realize the functions of analysis, calculation, storage and the like on the acquired data, and can realize real-time monitoring and control on the power system.

Zhejiang university began a relevant study on PMU in 1998, achieved some achievements in phase angle measurement and made a relevant prototype. In 2002, a PMU developed by Zhejiang university runs on a high-voltage network successfully, discrete Fourier transform is adopted for phase angle measurement of the PMU, and sampling synchronization in different places in the power grid is realized through GPS and PLL technologies.

The river-sea university starts to implement PMU development work in 2003, a GPS is adopted as a synchronous clock of the PMU developed by the river-sea university, and relevant experiments are carried out at a 500kV power transformation hub of a Henan power grid. And the system provides guarantee for a Wide Area Measurement System (WAMS) forming a Henan power grid.

A micro PMU device based on a DSP + main processor architecture is developed in 2005 by Beijing university of science and engineering, so that the miniaturization and the flexibility of a synchronous phasor measurement device are realized, and the current power grid development requirements are met.

In recent years, a lot of researches are carried out on phasor measurement methods by a plurality of domestic scholars, the performance of PMU hardware is improved, and the measured phasors can be more accurate by improving a phasor measurement algorithm. The technical specification of a real-time dynamic monitoring system of an electric power system, which is suitable for the application of the electric power system in China, is made by referring to the IEEE standard at the beginning of the 21 st century, and the technical requirements of synchrophasor measurement are explained in the specification.

There is much room for improvement in existing PMUs.

Disclosure of Invention

The application provides a power distribution network synchrophasor measurement device, which is specifically described below.

An embodiment provides a power distribution network synchronous phasor measurer, which comprises a main processor, a coprocessor, a first memory card, a second memory card, an Ethernet port, a time service module, an analog-to-digital converter and an edge calculation module, wherein the main processor is connected with the coprocessor;

the analog-to-digital converter is used for acquiring data at a power grid node and converting the data into digital signals from analog signals;

the time service module is used for acquiring time information of a satellite;

the coprocessor is respectively connected with the time service module, the module converter and the main processor; the coprocessor controls the analog-to-digital converter to acquire electric energy data through the time information, and calculates phasor data obtained by sampling by applying a discrete Fourier transform method; when the synchronous clock signal is lost or abnormal, the coprocessor realizes timekeeping for a period of time based on the clock of the coprocessor; when the power grid operating frequency is not fixed, the coprocessor acquires the power grid operating frequency in real time, corrects the power grid operating frequency, calculates synchronous sampling frequency, and calculates phasor data by applying a discrete Fourier transform method; the coprocessor transmits the phase data to the main processor after time stamping;

the main processor is connected with the first memory card and the second memory card respectively, the first memory card is used for storing short-term data, and the second memory card is used for storing fault recording information; the main processor uploads the phasor data received from the coprocessor through the Ethernet port, receives a control instruction issued by a superior, and downloads a corresponding control instruction to the coprocessor;

the edge calculation module is connected with the main processor and used for calling phasor data to calculate and returning a calculation result to the main processor; specifically, the edge calculation module calculates based on a distributed state estimation model, the distributed state estimation model is a prediction model based on integrated deep learning, the edge calculation module firstly decomposes input phasor data with time marks through an empirical mode decomposition algorithm to obtain sub-signals with different frequencies, then a deep circulation neural network is used for analyzing and predicting each sub-signal, and then the output obtained after each sub-signal is analyzed and predicted is integrated to complete the estimation of the state of the power grid; and/or the edge computing module is based on a fault early warning and relay protection model to carry out fault early warning and relay protection on elements needing to be protected in the power grid, wherein the fault early warning and relay protection model is constructed in the following way: step (1), a power load prediction algorithm is constructed by applying a deep circulation network according to phasor data acquired by a synchronous phasor measurer of a power distribution network; step (2), constructing a model based on a deep neural network for simulating an electric power system environment, wherein the model for simulating the electric power system environment is trained and verified by means of historical phasor data, current phasor data and phasor data predicted by an electric power load prediction algorithm; step (3), a relay protection optimization algorithm based on reinforcement learning is established, a reward function is defined by means of the model for simulating the power system environment established in the step (2), and an optimal strategy is obtained by applying the reinforcement learning algorithm so as to optimize a judgment standard and a judgment threshold of relay protection; and/or the edge calculation module performs calculation based on a fault location model, wherein the fault location model is constructed by: and performing regression fusion on the time domain information fault location algorithm and the frequency domain information fault location algorithm to obtain a fault location model.

In an embodiment, the power distribution network synchrophasor measurer further includes an IRIG-B interface and an RS485 interface, and the coprocessor can communicate with other devices through the IRIG-B interface and the RS485 interface to perform information interaction.

In an embodiment, the coprocessor is a digital signal processor.

In an embodiment, the time service module is a GPS/beidou time service module, and is capable of acquiring time information of the GPS/beidou.

In an embodiment, the power distribution network synchrophasor measurement unit further includes a USB interface, and the main processor communicates with other devices through the USB interface to perform information interaction.

In an embodiment, the power distribution network synchrophasor measurer further includes a display connected to the main processor for displaying information.

In an embodiment, after receiving an instruction for controlling the relay, which is downloaded by the main processor, the coprocessor controls the relay to be correspondingly turned on or off, and the switching information is uploaded back to the main processor.

In one embodiment, the edge calculation module comprises a GPU.

In an embodiment, the power distribution network synchronous phasor measurement unit further stores the phasor data with time stamps obtained by calculation, and stores information of the phasor data into a block chain, where the information of the phasor data includes one or more of type information of the phasor data, storage location information of the phasor data, version information of the phasor data, reading authority information of the phasor data, and read-write history information of the phasor data; when receiving a reading request of the stored phasor data, the main processor judges whether corresponding authority exists according to the reading authority information of the phasor data in the block chain, and when judging that the corresponding authority exists, the main processor allows reading of the phasor data stored by the power distribution network synchronous phasor measurer, otherwise, when judging that the corresponding authority does not exist, the main processor does not allow reading of the phasor data stored by the power distribution network synchronous phasor measurer.

According to the power distribution network synchronous phasor measurer of the embodiment, due to the fact that the edge calculation module is introduced, dependence on cloud can be eliminated, big data generated by the edge sensor do not need to be transmitted back to the cloud server, rapid operation can be conducted, and the requirement for data transmission is lowered.

Drawings

Fig. 1 is a schematic structural diagram of a power distribution network synchrophasor measurement apparatus according to an embodiment;

fig. 2 is a schematic structural diagram of a power distribution network synchrophasor measurement apparatus according to another embodiment;

fig. 3 is a schematic structural diagram of a synchrophasor measurement apparatus for a power distribution network according to yet another embodiment;

fig. 4 is a schematic structural diagram of a power distribution network synchrophasor measurement apparatus according to yet another embodiment;

FIG. 5 is a schematic diagram of an embodiment of an algorithm for state estimation;

FIG. 6 is a schematic diagram of an algorithm for state estimation according to another embodiment.

Detailed Description

The present invention will be described in further detail with reference to the following detailed description and accompanying drawings. Wherein like elements in different embodiments are numbered with like associated elements. In the following description, numerous specific details are set forth in order to provide a better understanding of the present application. However, one skilled in the art will readily recognize that some of the features may be omitted or replaced with other elements, materials, methods in different instances. In some instances, certain operations related to the present application have not been shown or described in this specification in order not to obscure the core of the present application with unnecessary detail, and it is not necessary for those skilled in the art to describe these operations in detail, so that they may be fully understood from the description in the specification and the general knowledge in the art.

Furthermore, the features, operations, or characteristics described in the specification may be combined in any suitable manner to form various embodiments. Also, the various steps or actions in the description of the methods may be transposed or transposed in order, as will be apparent to a person skilled in the art. Thus, the various sequences in the specification and drawings are for the purpose of describing certain embodiments only and are not intended to imply a required sequence unless otherwise indicated where such sequence must be followed.

The numbering of the components as such, e.g., "first", "second", etc., is used herein only to distinguish the objects as described, and does not have any sequential or technical meaning. The term "connected" and "coupled" when used in this application, unless otherwise indicated, includes both direct and indirect connections (couplings).

The characteristics of this application close combination future Distribution network, on the basis of fully understanding Phasor measuring Unit PMU's software and hardware framework and theory of operation, combine artificial intelligence edge calculation technique, develop the synchronous Phasor Measurement Unit DPMU (Distribution phase Measurement Unit) of Distribution network that the sexual valence relative altitude just is fit for functions such as load prediction, fault analysis or state estimation. This will be explained in detail below.

Referring to fig. 1, in some embodiments, the power distribution network synchronized phasor measurement apparatus includes a main processor 10, a coprocessor 12, a first memory card 14, a second memory card 16, an ethernet port 18, a time service module 20, an analog-to-digital converter 22, and an edge calculation module 24, which are described in detail below.

The analog-to-digital converter 22 is used for collecting data at a node of a power grid, and converting the data from an analog signal to a digital signal. For example, the analog-to-digital converter 22 collects the voltage and current at the grid node, and converts them into digital signals. In some examples, the analog-to-digital converter 22 may be an 8-way 16-bit high-speed analog-to-digital converter.

The time service module 20 is used for acquiring time information of the satellite. In some embodiments, the time service module 20 is a GPS/beidou time service module, and is capable of acquiring time information of the GPS/beidou. The GPS is abbreviated as Global Positioning System (GPS), and the time service module 20 can receive time information accurate to nanosecond level, which can be used for time service. The Beidou refers to the Chinese Beidou Satellite Navigation System (English name: beiDou Navigation Satellite System, BDS for short), which is a self-developed global Satellite Navigation System in China. The Beidou satellite navigation system consists of a space section, a ground section and a user section, can provide high-precision, high-reliability positioning, navigation and time service for various users all day long in the global scope, has short message communication capacity, and preliminarily has the capacity of regional navigation, positioning and time service, wherein the positioning precision is decimeter and centimeter level, the speed measurement precision is 0.2 meter/second, and the time service precision is 10 nanoseconds.

The coprocessor 12 is connected to a time service module 20, a module translator 22, and the main processor 10. The coprocessor 12 controls the analog-to-digital converter 22 to acquire the electric energy data through the time information acquired by the time service module 20, and calculates the phasor data obtained by sampling by applying a discrete fourier transform method. In some examples, coprocessor 12 implements a time-keeping based on its own clock when the synchronous clock signal is lost or abnormal. In some examples, when the grid operating frequency is not fixed, the co-processor 12 obtains the grid operating frequency in real time, corrects the grid operating frequency and calculates a synchronous sampling frequency, and then calculates phasor data by applying a discrete fourier transform method. The coprocessor 12 timestamps the data for transmission to the main processor 10. In some examples, the coprocessor 12 may be a digital signal processor DSP.

In some embodiments, referring to fig. 2, the power distribution network synchrophasor measurer further includes an IRIG-B interface and an RS485 interface, and the coprocessor 12 can communicate with other devices through the IRIG-B interface and the RS485 interface to perform information interaction.

The main processor 10 is connected to a first memory card 14 and a second memory card 16. The first memory card 14 is used to store short-term data, and the second memory card 16 is used to store fault recording information. The main processor 10 also uploads the phasor data received from the coprocessor 12 through the ethernet port 18, receives control commands issued by the upper level, and downloads corresponding control commands to the coprocessor 12. For example, when receiving a command for controlling the relay downloaded from the main processor 10, the coprocessor 12 controls the relay to be turned on or off accordingly, and uploads the switching information back to the main processor 10.

In some embodiments, referring to fig. 3, the power distribution network synchrophasor measurement apparatus further includes a USB interface, and the main processor 10 can communicate with other devices through the USB interface for information interaction.

In some embodiments, referring to fig. 4, the power distribution grid synchrophasor measurement apparatus further includes a display 26 connected to the main processor 10 for displaying information.

The edge calculation module 24 is connected to the main processor 10, and is configured to call the phasor data for calculation, and return the calculation result to the main processor. In some embodiments, edge calculation module 24 may include a GPU.

In some specific embodiments, the edge calculation module 24 performs calculation based on a distributed state estimation model, the distributed state estimation model is a prediction model based on integrated deep learning, the edge calculation module 24 decomposes the input phasor data with time stamps through an empirical mode decomposition algorithm to obtain sub-signals with different frequencies, analyzes and predicts each sub-signal by using a deep-layer recurrent neural network, and integrates the output obtained after each sub-signal is analyzed and predicted to complete the estimation of the power grid state.

In the aspect of a specific load prediction calculation or a distributed state estimation model, an accurate short-term power load time sequence prediction model is created in two research directions by using high-frequency data acquired by a DPMU in real time. The first is to use deep learning (deep learning) technology, whose basic principle is to use multiple nonlinear transformations to construct a complex network structure to extract high-level abstractions and features in the big data. The other direction is the integration of machine learning algorithms (Ensemble machine learning methods). Because a single prediction model is often unstable and cannot adapt to various conditions, the characteristics of time sequence signals are deeply researched through strategically combining different machine learning algorithms, various possible influence factors are fully considered, and a more accurate, more stable and efficient time sequence prediction model is constructed to predict the power load.

In some specific examples, a prediction model based on integrated deep learning may be constructed, where the prediction model decomposes input power load time series data through an empirical mode decomposition algorithm to obtain sub-signals with different frequencies, analyzes and predicts each sub-signal by using a deep-layer recurrent neural network, and finally integrates outputs obtained by analyzing and predicting each sub-signal to obtain a final result of load prediction. Referring to fig. 5, a flowchart of an algorithm of a prediction model or state estimation is shown, taking electricity data as power, in which a Time Series Signal (TSS) refers to a power Signal, and a Discrete Wavelet Transform (DWT) is performed on the Time Series Signal to obtain each item W1, W2, \ 8230;, wm; then, empirical mode decomposition is carried out on the W1, W2, \8230, wm respectively to obtain a plurality of sub-signals (such as IMF, R in the figure), specifically, empirical mode decomposition is carried out on the W1 to obtain

And R1; empirical mode decomposition of W1 results in->

And Rm; analyzing each sub-signal with Long-short term network (LSTM) to obtain respective prediction result, for example, based on the Long-short term network->

Is paired and/or matched>

Analyzed to get->

Using long-and short-term networks->

Is paired and/or matched>

Analyzed to get->

Using long and short term networks

Analysis of R1 gives->

Using long-and short-term networks->

Is paired and/or matched>

Analyzed to get->

Using long-and short-term networks->

Is paired and/or matched>

Analyzed to get->

Using long-and short-term networks->

Analyzing Rm to obtain

Then, the Prediction result of each sub-signal is used as the input of another long-short term neural network LSTM, and the final Prediction Result (PR) is obtained through training. In other examples, referring to fig. 6, the predicted result of each sub-signal and the corresponding Additional Features (AF) may be input to the long-term and short-term neural networks togetherAnd connecting with LSTM to train to obtain the final prediction result. Additional features here may be currents, voltages and phase angles or phase angle differences etc. Taking the phase angle as an example: the power is divided into active power and reactive power, and the respective occupation ratios are determined by the phase angle difference of the voltage and the current, so the current, the voltage and the respective phase angle can provide more detailed information for power calculation, and are also helpful for power prediction in the model.

The load prediction model or the distributed state estimation model is exemplified by LSTM, and actually, there are many timing prediction algorithms that can be used, including traditional statistical methods (such as linear regression, ARIMA, etc.), classical machine learning algorithms (neural networks, support vector machines, random forests, etc.), and deep learning algorithms (convolutional neural networks, cyclic neural networks, etc.). Aiming at the application scene of edge calculation, an ensemble learning algorithm which is based on a simple model and integrates all models to output and obtain an accurate prediction result in a policy mode can be preferably considered, the calculation speed and the prediction accuracy can be comprehensively considered, and the calculation performance requirement on edge calculation equipment is reduced.

In some specific embodiments, the edge computing module 24 performs fault early warning and relay protection on an element to be protected in the power grid based on a fault early warning and relay protection model, where the fault early warning and relay protection model is constructed by: step (1), a power load prediction algorithm is constructed by applying a deep circulation network according to phasor data acquired by a synchronous phasor measurer of a power distribution network; step (2), constructing a model based on a deep neural network for simulating an electric power system environment, wherein the model for simulating the electric power system environment is trained and verified by means of historical phasor data, current phasor data and phasor data predicted by an electric power load prediction algorithm; and (3) constructing a relay protection optimization algorithm based on reinforcement learning, defining a reward function by depending on the model for simulating the power system environment constructed in the step (2), and applying the reinforcement learning algorithm to obtain an optimal strategy so as to optimize a judgment standard and a judgment threshold of relay protection. In some embodiments, step (3) specifically includes: establishing a generation countermeasure network for unsupervised learning by means of phasor data acquired by a synchronous phasor measurer of the power distribution network, so that a generation model learns the rule and the mode of the environmental operation of the power system in a normal mode, and simultaneously generates simulation data to help a judgment model to learn the abnormal condition of an element to be protected; through continuous iteration, the generation model and the judgment model compete with each other to obtain a substitution model of the model for simulating the power system environment and a classifier capable of judging the abnormal condition of the element to be protected, and the classifier is used for analyzing the prediction result of the power load prediction algorithm in the step (1) to judge the abnormal condition of the element to be protected; based on the power load prediction algorithm and the generation countermeasure network, a simulated operation environment of the element to be protected in the power grid is constructed, after the simulated operation environment of the element to be protected in the power grid is constructed, a reward function or an optimization target is defined, and a judgment standard and a judgment threshold value of relay protection are optimized by applying a reinforcement learning algorithm.

In summary, for the application scenario of the edge calculation, in some examples, DPMU may be used to acquire the relevant operating parameters of the protected element, and meanwhile, artificial intelligent models such as time sequence prediction, environmental simulation, fault resolution, and the like are constructed, and based on the conventional relay protection strategy, the relevant threshold is continuously optimized and iterated, so as to finally obtain the optimal protection strategy. In some examples, the marginal intelligent relay protection scheme of the DPMU has the advantages of high calculation speed, low data transmission requirement, high safety, and low false alarm rate and misoperation rate.

In some specific embodiments, the edge calculation module 24 performs the calculation based on a fault localization model, wherein the fault localization model is constructed by: and performing regression fusion on the time domain information fault location algorithm and the frequency domain information fault location algorithm to obtain a fault location model. In some specific embodiments, performing regression fusion on the time domain information fault location algorithm and the frequency domain information fault location algorithm to obtain a fault location model, includes: a machine learning method is adopted, a training set is formed by utilizing simulation data, a machine learning algorithm is trained to obtain a regression model, and when a power distribution network fails, multi-criterion fusion is achieved based on the regression model. In other specific embodiments, performing regression fusion on the time domain information fault location algorithm and the frequency domain information fault location algorithm to obtain a fault location model includes: and constructing a selection model of the multi-fault positioning result by using an ensemble learning method, training the selection model of the multi-fault positioning result by using simulation data as a training set, and selecting a result with the minimum error from the multi-fault positioning result by using the selection model as output when the power distribution network has a fault.

One inventive point of the present invention is to introduce the edge calculation module 24, and the DPMU in some examples of the present invention may acquire the values and phase angles of the vectors in the power grid in real time at a high frequency, so as to provide a powerful hardware support for the edge calculation. The edge computing technology can get rid of dependence on cloud, large data generated by the edge sensor does not need to be transmitted back to a cloud server, rapid operation can be carried out, the requirement on data transmission is lowered, and meanwhile, the operation speed is also improved. In some examples, edge computation module 24 may run locally on the DPMU using low power small GPU chips with complex matrix operations for artificial intelligence correlation and state estimation. The edge calculation module 24 cooperates with the main processor 10 to call data in the DPMU and return the calculation result.

In some embodiments, the power distribution network synchrophasor measurer further stores the phasor data with the time stamp obtained through calculation, and stores information of the phasor data into the block chain, wherein the information of the phasor data includes one or more of type information of the phasor data, storage location information of the phasor data, version information of the phasor data, reading authority information of the phasor data, and reading and writing history information of the phasor data; when receiving a reading request of the stored phasor data, the main processor judges whether corresponding authority exists according to the reading authority information of the phasor data in the block chain, if so, the main processor allows reading of the phasor data stored by the power distribution network synchronous phasor measurer, otherwise, if not, the main processor does not allow reading of the phasor data stored by the power distribution network synchronous phasor measurer

The DPMU in some embodiments has microsecond resolution and ultra-high precision required by power distribution network applications, achieves ultra-high precision synchronous phasor measurement, conforms to the IEEE c37.118.2-2011 standard of synchronous phasor measurement devices, and can be used for real-time control applications. DPMU in some embodiments has ultra-accurate phasor measurements: TVE 0.01%. The DPMUs in some embodiments are capable of simultaneously recording transmit synchrophasor data. DPMU in some embodiments is capable of fast recording/streaming rates: 100 times/sec at 50Hz and 120 times/sec at 60 Hz. Voltage and current phasors collected by DPMU in some embodiments: frequency, active/reactive power, power factor. The DPMU in some embodiments conforms to IEEE c37.118.1-2011 and c37.118.2-2011. DPMU in some embodiments is fully compatible with OpenPDC. The DPMU internal memory in some embodiments may hold 30 days of phasor measurement data. In some embodiments, the DPMU may have its data records downloaded via FTP. The DPMU in some embodiments may be configured and firmware updated via FTP or web pages (HTTP). The DPMU in some embodiments supports connections through PT and CT. The DPMU in some embodiments supports 5G network, beidou and GPS time synchronization. The DPMU in some embodiments has low power consumption edge computing capability, supporting artificial intelligence model computing.

Reference is made herein to various exemplary embodiments. However, those skilled in the art will recognize that changes and modifications may be made to the exemplary embodiments without departing from the scope hereof. For example, the various operational steps, as well as the components used to perform the operational steps, may be implemented in differing ways depending upon the particular application or consideration of any number of cost functions associated with operation of the system (e.g., one or more steps may be deleted, modified or incorporated into other steps).

While the principles herein have been illustrated in various embodiments, many modifications of structure, arrangement, proportions, elements, materials, and components particularly adapted to specific environments and operative requirements may be employed without departing from the principles and scope of the present disclosure. The above modifications and other changes or modifications are intended to be included within the scope of this document.

The foregoing detailed description has been described with reference to various embodiments. However, one skilled in the art will recognize that various modifications and changes may be made without departing from the scope of the present disclosure. Accordingly, the disclosure is to be considered in an illustrative and not a restrictive sense, and all such modifications are intended to be included within the scope thereof. Also, advantages, other advantages, and solutions to problems have been described above with regard to various embodiments. However, the benefits, advantages, solutions to problems, and any element(s) that may cause any element(s) to occur or become more pronounced are not to be construed as a critical, required, or essential feature or element of any or all the claims. As used herein, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, system, article, or apparatus. Furthermore, the term "coupled," and any other variation thereof, as used herein, refers to a physical connection, an electrical connection, a magnetic connection, an optical connection, a communicative connection, a functional connection, and/or any other connection.

Those skilled in the art will recognize that many changes may be made to the details of the above-described embodiments without departing from the underlying principles of the invention. Accordingly, the scope of the invention should be determined only by the claims.

Claims

1. A synchronous phasor measurer for a power distribution network is characterized by comprising a main processor, a coprocessor, a first memory card, a second memory card, an Ethernet port, a time service module, an analog-to-digital converter and an edge calculation module;

the time service module is used for acquiring time information of a satellite;

the coprocessor is respectively connected with the time service module, the analog-to-digital converter and the main processor; the coprocessor controls the analog-to-digital converter to acquire electric energy data through the time information, and calculates phasor data obtained by sampling by applying a discrete Fourier transform method; when the synchronous clock signal is lost or abnormal, the coprocessor realizes timekeeping for a period of time based on the clock of the coprocessor; when the power grid operating frequency is not fixed, the coprocessor acquires the power grid operating frequency in real time, corrects the power grid operating frequency, calculates synchronous sampling frequency, and calculates phasor data by applying a discrete Fourier transform method; the coprocessor transmits the phase data to the main processor after time stamping;

the edge calculation module is connected with the main processor and used for calling phasor data to calculate and returning a calculation result to the main processor; the edge calculation module carries out calculation based on a distributed state estimation model, the distributed state estimation model is a prediction model based on integrated deep learning, the edge calculation module firstly decomposes input phasor data with time marks through an empirical mode decomposition algorithm to obtain sub-signals with different frequencies, then a deep circulating neural network is used for analyzing and predicting each sub-signal, and then the output obtained after each sub-signal is analyzed and predicted is integrated to complete the estimation of the state of the power grid; firstly, discrete wavelet transform is carried out on the time sequence signal to obtain W1, W2, \ 8230;, wm; then, the items W1, W2, \ 8230, wm are respectively processed with empirical mode decomposition to obtain a plurality of sub-signals, wherein, W1 is processed with empirical mode decomposition to obtain

And R1; empirical mode decomposition of Wm results in>

And Rm; then, the long-term and short-term networks are applied to analyze each sub-signal to obtain respective predictionResult, in which a long-and short-term network is applied->

Is paired and/or matched>

Analyzed to get->

Using long-and short-term networks->

Is paired and/or matched>

Analysis results in>

Using long and short term networks>

Analysis of R1 results in +>

Using long-and short-term networks->

Is paired and/or matched>

Analyzed to get->

Using long and short term networks

Is paired and/or matched>

Analyzed to get->

Using long-and short-term networks->

Analysis of Rm gives->

And then the prediction result of each sub-signal is used as the input of another long-short term neural network LSTM, and the final prediction result is obtained through training.

2. The synchronized phasor measurer for power distribution networks of claim 1, further comprising an IRIG-B interface and an RS485 interface, wherein the coprocessor can communicate with other devices through the IRIG-B interface and the RS485 interface to perform information interaction.

3. The power distribution network synchrophasor measurer according to claim 1 or 2, wherein said coprocessor is a digital signal processor.

4. The power distribution network synchrophasor measurer of claim 1, wherein the time service module is a GPS/beidou time service module capable of acquiring time information of GPS/beidou.

5. The synchrophasor measurement instrument for power distribution network of claim 1, further comprising a USB interface, wherein said main processor communicates with other devices through said USB interface for information interaction.

6. The power distribution network synchrophasor measurer of claim 1, further comprising a display connected to said main processor for displaying information.

7. The synchrophasor measurer for power distribution network of claim 1 wherein, when receiving the command of controlling the relay downloaded by the main processor, the coprocessor controls the relay to be turned on or off correspondingly and uploads the switching information back to the main processor.

8. The power distribution network synchrophasor measurer according to any of claims 1 to 7, wherein said edge calculation module comprises a GPU.

9. The power distribution network synchrophasor measurer of claim 1, further storing the phasor data with a time stamp calculated by the power distribution network synchrophasor measurer, and storing information of the phasor data to a block chain, wherein the information of the phasor data includes one or more of type information of the phasor data, storage location information of the phasor data, version information of the phasor data, reading authority information of the phasor data, and reading and writing history information of the phasor data; when receiving a reading request of the stored phasor data, the main processor judges whether corresponding authority exists according to the reading authority information of the phasor data in the block chain, and when judging that the corresponding authority exists, the main processor allows reading of the phasor data stored by the power distribution network synchronous phasor measurer, otherwise, when judging that the corresponding authority does not exist, the main processor does not allow reading of the phasor data stored by the power distribution network synchronous phasor measurer.