CN114091750A - Power grid load abnormity prediction method, system and storage medium - Google Patents

Power grid load abnormity prediction method, system and storage medium Download PDF

Info

Publication number
CN114091750A
CN114091750A CN202111383136.1A CN202111383136A CN114091750A CN 114091750 A CN114091750 A CN 114091750A CN 202111383136 A CN202111383136 A CN 202111383136A CN 114091750 A CN114091750 A CN 114091750A
Authority
CN
China
Prior art keywords
model
sequence
prediction
load
power grid
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111383136.1A
Other languages
Chinese (zh)
Inventor
余芸
明哲
萧展辉
甘杉
甘莹
邓丽娟
李文俊
马赟
冯志宏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Southern Power Grid Digital Grid Research Institute Co Ltd
Original Assignee
Southern Power Grid Digital Grid Research Institute Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Southern Power Grid Digital Grid Research Institute Co Ltd filed Critical Southern Power Grid Digital Grid Research Institute Co Ltd
Priority to CN202111383136.1A priority Critical patent/CN114091750A/en
Publication of CN114091750A publication Critical patent/CN114091750A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Marketing (AREA)
  • Biophysics (AREA)
  • General Business, Economics & Management (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Development Economics (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • Primary Health Care (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The application discloses a power grid load abnormity prediction method, a system and a storage medium, which relate to an intelligent power grid and artificial intelligence and comprise the following steps: acquiring power grid load sequence data; inputting the power grid load sequence data into a pre-trained load abnormity prediction model for prediction to obtain a prediction result of the next time period; comparing the prediction result of the next time period with the early warning value, and judging whether an abnormality occurs; the load abnormity prediction model comprises a first sequence model and a second sequence model, and the prediction result of the next time period is determined according to the weighted result of the first sequence model and the second sequence model on the prediction result of the power grid load sequence data. The method and the device can predict the future load state in advance, judge whether the load is abnormal or not according to the future load state, and can process in time.

Description

Power grid load abnormity prediction method, system and storage medium
Technical Field
The application relates to a smart grid and artificial intelligence, in particular to a method, a system and a storage medium for predicting load abnormity of a grid.
Background
With the development of power technology, more and more new technologies are fused into a power grid to form an intelligent power grid. Smart grids put higher demands on the performance of the grid. Such as visualization, real-time visualization of grid data.
The method is characterized in that faults occur in the past power grid due to local overload, the faults occur in the past, and measures are taken after the faults occur. Such an approach has not been able to meet the needs of the smart grid. In the traditional scheme, some abnormal conditions can be found through a real-time data and threshold value comparison mode, but early warning can be delayed in time. Therefore, a new method is needed to predict the load abnormality in advance.
Disclosure of Invention
The present invention is directed to solving at least one of the problems of the prior art. Therefore, the invention provides a power grid load abnormity prediction method, a power grid load abnormity prediction system and a storage medium, which can predict the load abnormity condition in advance.
The embodiment of the application provides:
a power grid load abnormity prediction method comprises the following steps:
acquiring power grid load sequence data, wherein the power grid load sequence data is data to be predicted, the power grid load sequence data is composed of a plurality of statistical points, each statistical point represents power grid load statistics of a time period on a time sequence, and the time periods corresponding to the statistical points are sequentially in end connection and are not overlapped;
inputting the power grid load sequence data into a pre-trained load abnormity prediction model for prediction to obtain a prediction result of the next time period;
comparing the prediction result of the next time period with the early warning value, and judging whether an abnormality occurs;
the load abnormity prediction model comprises a first sequence model and a second sequence model, and the prediction result of the next time period is determined according to the weighted result of the first sequence model and the second sequence model on the prediction result of the power grid load sequence data.
In some embodiments, the training data for the first and second sequence models is obtained by:
the monitoring equipment of the power grid transmission node is configured to mark and upload power grid load data before and after load abnormity occurs when the load abnormity is detected, wherein the time before and after the load abnormity occurs refers to a time period with a preset length taking a time node of the load occurrence as a central point;
and constructing a training set according to the abnormal data uploaded by each detection device.
In some embodiments, the first sequence model is different from the second sequence model.
In some embodiments, the first sequence model is an LSTM model and the second sequence model is a GRU model;
and the prediction result of the next time period is obtained by averaging a first prediction result of the LSTM model on the power grid load sequence data and a second prediction result of the GRU model on the power grid load sequence data.
In some embodiments, the grid load sequence data is represented by a vector V ═ { x1, x2, x3, … …, xn }, n being a positive integer, each element in the vector representing a grid load statistic per unit time.
In some embodiments, the LSTM model is comprised of a plurality of model cells, each of the model cells including a forgetting gate, an input gate, and an output gate;
wherein, the 1 st model unit in the LSTM model generates output data according to the input data at the 1 st moment, and the mth model unit generates output data according to the output data set of the previous unit and the input data at the mth moment.
On the other hand, the embodiment discloses a power grid load abnormality prediction system, which includes:
the system comprises an acquisition unit, a prediction unit and a prediction unit, wherein the acquisition unit is used for acquiring power grid load sequence data, the power grid load sequence data is data to be predicted, the power grid load sequence data is composed of a plurality of statistic points, each statistic point represents power grid load statistic of a time period on a time sequence, and the time periods corresponding to the statistic points are sequentially connected in an end-to-end manner and are not overlapped;
the prediction unit is used for inputting the power grid load sequence data into a pre-trained load abnormity prediction model for prediction to obtain a prediction result of the next time period;
the judging unit is used for comparing the prediction result of the next time period with the early warning value and judging whether the abnormity occurs or not;
the load abnormity prediction model comprises a first sequence model and a second sequence model, and the prediction result of the next time period is determined according to the weighted result of the first sequence model and the second sequence model on the prediction result of the power grid load sequence data.
In some embodiments, the first sequence model is an LSTM model and the second sequence model is a GRU model;
and the prediction result of the next time period is obtained by averaging a first prediction result of the LSTM model on the power grid load sequence data and a second prediction result of the GRU model on the power grid load sequence data.
In some embodiments, the LSTM model is comprised of a plurality of model cells, each of the model cells including a forgetting gate, an input gate, and an output gate;
wherein, the 1 st model unit in the LSTM model generates output data according to the input data at the 1 st moment, and the mth model unit generates output data according to the output data set of the previous unit and the input data at the mth moment.
In another aspect, the present embodiment discloses a computer-readable storage medium, which stores a program, and when the program is executed by a processor, the method for predicting grid load abnormality is implemented.
According to the embodiment of the application, the current power grid load sequence data is obtained, and then the power grid load sequence data is input into a pre-trained load abnormity prediction model for prediction, so that the prediction result of the next time period is obtained; finally, comparing the prediction result of the next time period with the early warning value, and judging whether an abnormality occurs; the load condition of the next time period can be predicted by using the current data, so that the possible load abnormity problem can be found in advance; the load abnormity prediction model is predicted by two different submodels, so that the prediction accuracy can be improved, abnormity early warning can be realized in time, and managers can be helped to make decisions.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart of a method for predicting load abnormality of a power grid according to an embodiment of the present disclosure;
fig. 2 is a schematic structural diagram of a load anomaly prediction model provided in an embodiment of the present application;
fig. 3 is a schematic diagram of a model element of an LSTM network provided in an embodiment of the present application.
Detailed Description
In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below through embodiments with reference to the accompanying drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description of the present invention, the meaning of a plurality is one or more, the meaning of a plurality is two or more, and the above, below, exceeding, etc. are understood as excluding the present numbers, and the above, below, within, etc. are understood as including the present numbers. If the first and second are described for the purpose of distinguishing technical features, they are not to be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
In the description of the present invention, unless otherwise explicitly defined, terms such as set, etc. should be broadly construed, and those skilled in the art can reasonably determine the specific meanings of the above terms in the present invention in combination with the detailed contents of the technical solutions.
In the description of the present invention, reference to the description of the terms "one embodiment," "some embodiments," "an illustrative embodiment," "an example," "a specific example," or "some examples," etc., means that a particular feature or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The noun explains: the total load of the power system is the sum of total power consumed by all the electric equipment in the system; adding the power consumed by the industrial, agricultural, post and telecommunications, traffic, municipal, commercial and urban and rural residents to obtain the comprehensive power load of the power system; the power of the comprehensive power load plus the network loss is the power to be supplied by each power plant in the system, and is called the power supply load (power supply amount) of the power system; the power supply load plus the power consumed by each power plant (i.e., the service power) is the power that each generator in the system should generate, and is called the power generation load (power generation amount) of the system. It should be understood that in the present application, the grid load may refer to the supply and demand relationship local to the grid.
The intelligent power grid: the intelligent power grid is intelligentized and is also called as 'power grid 2.0', the power grid is established on the basis of an integrated high-speed bidirectional communication network, and the power grid is reliably, safely, economically, efficiently, environmentally-friendly and safe to use through the application of advanced sensing and measuring technology, advanced equipment technology, advanced control method and advanced decision support system technology. In the application, the artificial intelligence technology is fused into the intelligent power grid, and the intellectualization of part of functions of the power grid is realized.
Artificial intelligence: artificial Intelligence, abbreviated as AI. The method is a new technical science for researching and developing theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can react in a manner similar to human intelligence, a field of research that includes robotics, language recognition, image recognition, natural language processing, and expert systems, among others. Since the birth of artificial intelligence, theories and technologies become mature day by day, and application fields are expanded continuously, so that science and technology products brought by the artificial intelligence in the future can be assumed to be 'containers' of human intelligence. The artificial intelligence can simulate the information process of human consciousness and thinking. Artificial intelligence is not human intelligence, but can think like a human, and can also exceed human intelligence. In the present application, artificial intelligence techniques are involved, in particular temporal sequence prediction techniques.
A Recurrent Neural Network (RNN) is a type of Recurrent Neural Network (Recurrent Neural Network) in which sequence data is input, recursion is performed in the direction of evolution of the sequence, and all nodes (Recurrent units) are connected in a chain. In 2014, a Gated recycling Unit networks (GRUs) was proposed, which is a friendship RNN.
The Long Short-Term Memory network (LSTM) is a time-cycle neural network, which is specially designed to solve the Long-Term dependence problem of the general RNN (cyclic neural network), and all RNNs have a chain form of repeated neural network modules.
Referring to fig. 1, a method for predicting load abnormality of a power grid may be used to predict a load condition in a future period of time, and determine whether a load abnormality exists, so that an operation and maintenance worker can command and process the load abnormality in time, in this embodiment, the method includes the following steps:
s100, acquiring power grid load sequence data, wherein the power grid load sequence data is data to be predicted, the power grid load sequence data is composed of a plurality of statistic points, each statistic point represents power grid load statistic of a time period on a time sequence, and the time periods corresponding to the statistic points are sequentially connected in an end-to-end mode and are not overlapped.
In some embodiments, the grid load sequence data is represented by a vector V ═ { x1, x2, x3, … …, xn }, n being a positive integer, each element in the vector representing a grid load statistic per unit time. The unit time may be 5 minutes, 10 minutes, or the like, for example, x1 represents the load state for the period of time from 10 o 'clock to 10 o' clock 05 minutes; x2 represents the load condition from 10 o 'clock 05 to 10 o' clock 10; x2 represents the load condition from 10 o 'clock 10 to 10 o' clock 15, and so on. The grid load may refer to a percentage of the load. And in a unit time, the maximum load capacity is taken as the corresponding load state in the unit time according to the actual average load capacity.
And S200, inputting the power grid load sequence data into a pre-trained load abnormity prediction model for prediction to obtain a prediction result of the next time period.
For example, the input data is data load sequence data of 10 to 11 points, which is counted every 5 minutes. The example outputs a load status of 11 points to 11 points and 5 points. Corresponding to a situation where five minutes into the future is predicted in advance. Therefore, operation and maintenance personnel can be helped to find problems in time.
S300, comparing the prediction result of the next time period with the early warning value, and judging whether abnormality occurs. In the step, a threshold comparison mode is still adopted, when the prediction result exceeds the threshold, the fact that the prediction result is abnormal can be judged, and operation and maintenance personnel can take measures in time to avoid the abnormal occurrence.
The load abnormity prediction model comprises a first sequence model and a second sequence model, and the prediction result of the next time period is determined according to the weighted result of the first sequence model and the second sequence model on the prediction result of the power grid load sequence data.
In this embodiment, the first sequence model and the second sequence model are different models, and the differences include different model structures, different training data, different satisfied performances, and the like. In the embodiments, two different models are adopted for prediction, and the final result is determined according to the prediction results of the two models, so that the problem that the accuracy rate and the recall rate of a single model cannot be met at the same time is avoided, different models are used for making up each other, and the prediction accuracy is improved.
According to the embodiment, the current power grid load sequence data is obtained, and then the power grid load sequence data is input into a pre-trained load abnormity prediction model for prediction, so that the prediction result of the next time period is obtained; finally, comparing the prediction result of the next time period with the early warning value, and judging whether an abnormality occurs; the load condition of the next time period can be predicted by using the current data, so that the possible load abnormal problem can be found in advance; the load abnormity prediction model is predicted by two different submodels, so that the prediction accuracy can be improved, abnormity early warning can be realized in time, and management personnel can be helped to make decisions.
In some embodiments, the training data for the first and second sequence models is obtained by:
the monitoring equipment of the power grid transmission node is configured to mark and upload power grid load data before and after load abnormity occurs when the load abnormity is detected, wherein the time before and after the load abnormity occurs refers to a time period with a preset length taking a time node of the load occurrence as a central point;
and constructing a training set according to the abnormal data uploaded by each detection device.
In the above embodiment, the labor cost can be greatly reduced by determining the abnormality according to the device deployed in the terminal and automatically generating the training data according to the abnormality. And for terminal equipment, the number of sensors is relatively large, the data is more accurate compared with that observed manually when abnormity is judged, data generation is real-time, and data collection can be directly carried out in the operation process of a power grid. Of course, since the situation of different grids is different (for example, the situation of two parts is different), the data needs to be normalized before being processed. It will be appreciated that normalization is also required when prediction is performed.
In some embodiments, the first sequence model is different from the second sequence model. It will be appreciated that the first and second sequence models may take the form of different sequence models, including RNN networks, LSTM models, GRU models, and so on.
Referring to fig. 2, in some embodiments, the first sequence model is an LSTM model and the second sequence model is a GRU model;
and the prediction result of the next time period is obtained by averaging a first prediction result of the LSTM model on the power grid load sequence data and a second prediction result of the GRU model on the power grid load sequence data.
In some embodiments, the LSTM model is comprised of a plurality of model cells, each of the model cells including a forgetting gate, an input gate, and an output gate;
wherein, the 1 st model unit in the LSTM model generates output data according to the input data at the 1 st moment, and the mth model unit generates output data according to the output data set of the previous unit and the input data at the mth moment.
Referring to fig. 3, the LSTM network is a neural network composed of a plurality of repetitive structures, and fig. 3 is a schematic diagram of one of the structures (model unit), and the model unit operates according to the following principle: first, the model unit has an output and a cell state at each time, and in FIG. 3, the current time is represented by time t, and the output is h as showntThe cell state is Ct. For the current time, outputting h of the last timet-1As input x of the cell state at the current moment, in conjunction with the input at the current moment as a wholetAnd is also the input to control three gates (gates).
During the treatment: firstly, the output at the previous moment passes through a forgetting Gate (Forget Gate), if the output result of the forgetting Gate is close to 0, the output at the previous moment is forgotten as much as possible, and the close to 1 indicates that the output at the previous moment is memorized as much as possible; then, the remaining portion of the previous state in the cell after the forgetting Gate continues forward, and the Input at the current time is added through an Input Gate (Input Gate) which represents how much Input is passed, and the Input at the current time is processed through the Input Gate (the Input is subjected to tanh operation before entering the Input Gate, so that the value range is compressed to [ -1,1 []In range) and previous last cell state Ct-1The residue of (a) is added up to be the cell state at this time. Thus, the cell state is updated from the last time to the present time. And cell state C after tanh calculationtMultiplied by an Output Gate (Output Gate), the result is the Output h at the current timet. The long-short term memory neural network can overcome the problem of gradient disappearance when a simple RNN neural network processes a long-term dependent time sequence, and has excellent prediction performance on a time sequence prediction task.
Unlike the three-gated mode of LSTM, the GRU model has only two gates, namely reset gate (reset gate) and update gate (update gate), although it also has two gates. The reset gate controls whether to reset, i.e. how much to erase the previous state, as the name implies; the update gate indicates how much to update the current hidden layer with the current input
The basic principle of the GRU is that firstly, two gates are generated by using x (t) and h (t-1), then the state of the last moment is multiplied by a reset gate, then whether reset is needed or how much reset is needed is judged, then, the gate is spliced with a new input x, the network is passed and activated by tanh, an implicit variable hat { h _ t } of the current input is formed, then, the h of the last moment and the h of the current input are linearly combined, the sum of the weights of the h and the h of the current input is 1, and the weight of the current input is the output of the update gate, and the update degree is represented. Note that h is just one variable, so at each instant, including the last linear combination, h is updating itself with the previous and current alternative answers. Similarly, it will be appreciated that LSTM, whose forgetting gate functions similarly to a reset, and whose input gate is similar to an update gate, except that LSTM also controls the output of the current state, i.e., the function of the output gate, is not available to the GRU.
The embodiment discloses a power grid load abnormity prediction system, which comprises:
the system comprises an acquisition unit, a prediction unit and a prediction unit, wherein the acquisition unit is used for acquiring power grid load sequence data, the power grid load sequence data is data to be predicted, the power grid load sequence data is composed of a plurality of statistic points, each statistic point represents power grid load statistic of a time period on a time sequence, and the time periods corresponding to the statistic points are sequentially connected in an end-to-end manner and are not overlapped;
the prediction unit is used for inputting the power grid load sequence data into a pre-trained load abnormity prediction model for prediction to obtain a prediction result of the next time period;
the judging unit is used for comparing the prediction result of the next time period with the early warning value and judging whether abnormity occurs or not;
the load abnormity prediction model comprises a first sequence model and a second sequence model, and the prediction result of the next time period is determined according to the weighted result of the first sequence model and the second sequence model on the prediction result of the power grid load sequence data.
According to the embodiment, the current power grid load sequence data is obtained, and then the power grid load sequence data is input into a pre-trained load abnormity prediction model for prediction, so that the prediction result of the next time period is obtained; finally, comparing the prediction result of the next time period with the early warning value, and judging whether abnormity occurs or not; the load condition of the next time period can be predicted by using the current data, so that the possible load abnormal problem can be found in advance; the load abnormity prediction model is predicted by two different submodels, so that the prediction accuracy can be improved, abnormity early warning can be realized in time, and management personnel can be helped to make decisions.
In some embodiments, the first sequence model is an LSTM model and the second sequence model is a GRU model;
and the prediction result of the next time period is obtained by averaging a first prediction result of the LSTM model on the power grid load sequence data and a second prediction result of the GRU model on the power grid load sequence data.
In some embodiments, the LSTM model is comprised of a plurality of model cells, each of the model cells including a forgetting gate, an input gate, and an output gate;
wherein, the 1 st model unit in the LSTM model generates output data according to the input data at the 1 st moment, and the mth model unit generates output data according to the output data set of the previous unit and the input data at the mth moment.
The embodiment of the present application further provides a computer-readable storage medium, where the computer-readable storage medium is used for storing a program code, and the program code is used for executing the power grid load abnormality prediction method in each of the above embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one logical division, and other divisions may be realized in practice, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. The storage medium may include: a U-disk, a portable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other media capable of storing program codes.
The step numbers in the above method embodiments are set for convenience of illustration only, the order between the steps is not limited at all, and the execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present application and the technical principles employed. It will be understood by those skilled in the art that the present application is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the application. Therefore, although the present application has been described in more detail with reference to the above embodiments, the present application is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims (10)

1. The method for predicting the load abnormity of the power grid is characterized by comprising the following steps of:
acquiring power grid load sequence data, wherein the power grid load sequence data is data to be predicted, the power grid load sequence data is composed of a plurality of statistical points, each statistical point represents power grid load statistics of a time period on a time sequence, and the time periods corresponding to the statistical points are sequentially in end connection and are not overlapped;
inputting the power grid load sequence data into a pre-trained load abnormity prediction model for prediction to obtain a prediction result of the next time period;
comparing the prediction result of the next time period with the early warning value, and judging whether an abnormality occurs;
the load abnormity prediction model comprises a first sequence model and a second sequence model, and the prediction result of the next time period is determined according to the weighted result of the first sequence model and the second sequence model on the prediction result of the power grid load sequence data.
2. The grid load anomaly prediction method according to claim 1, wherein the training data of the first sequence model and the second sequence model are obtained by:
the monitoring equipment of the power grid transmission node is configured to mark and upload power grid load data before and after load abnormity occurs when the load abnormity is detected, wherein the time before and after the load abnormity occurs refers to a time period with a preset length taking a time node of the load occurrence as a central point;
and constructing a training set according to the abnormal data uploaded by each detection device.
3. The grid load anomaly prediction method according to claim 2, characterized in that the first sequence model is different from the second sequence model.
4. The grid load anomaly prediction method according to claim 3, characterized in that the first sequence model is an LSTM model and the second sequence model is a GRU model;
and the prediction result of the next time period is obtained by averaging a first prediction result of the LSTM model on the power grid load sequence data and a second prediction result of the GRU model on the power grid load sequence data.
5. The grid load abnormality prediction method according to claim 1, wherein the grid load sequence data is represented by a vector V ═ { x1, x2, x3, … …, xn }, where n is a positive integer, and each element in the vector represents a grid load statistic per unit time.
6. The grid load anomaly prediction method according to claim 4, characterized in that the LSTM model is composed of a plurality of model units, each model unit comprises a forgetting gate, an input gate and an output gate;
wherein, the 1 st model unit in the LSTM model generates output data according to the input data at the 1 st moment, and the mth model unit generates output data according to the output data set of the previous unit and the input data at the mth moment.
7. A grid load anomaly prediction system, comprising:
the system comprises an acquisition unit, a prediction unit and a prediction unit, wherein the acquisition unit is used for acquiring power grid load sequence data, the power grid load sequence data is data to be predicted, the power grid load sequence data is composed of a plurality of statistic points, each statistic point represents power grid load statistic of a time period on a time sequence, and the time periods corresponding to the statistic points are sequentially connected in an end-to-end manner and are not overlapped;
the prediction unit is used for inputting the power grid load sequence data into a pre-trained load abnormity prediction model for prediction to obtain a prediction result of the next time period;
the judging unit is used for comparing the prediction result of the next time period with the early warning value and judging whether abnormity occurs or not;
the load abnormity prediction model comprises a first sequence model and a second sequence model, and the prediction result of the next time period is determined according to the weighted result of the first sequence model and the second sequence model on the prediction result of the power grid load sequence data.
8. The grid load anomaly prediction system according to claim 7, characterized in that the first sequence model is an LSTM model and the second sequence model is a GRU model;
and the prediction result of the next time period is obtained by averaging a first prediction result of the LSTM model on the power grid load sequence data and a second prediction result of the GRU model on the power grid load sequence data.
9. The grid load anomaly prediction system according to claim 8, characterized in that the LSTM model is composed of a plurality of model units, each of which comprises a forgetting gate, an input gate and an output gate;
wherein, the 1 st model unit in the LSTM model generates output data according to the input data at the 1 st moment, and the mth model unit generates output data according to the output data set of the previous unit and the input data at the mth moment.
10. A computer-readable storage medium, characterized in that it stores a program which, when executed by a processor, implements the grid load anomaly prediction method according to any one of claims 1 to 7.
CN202111383136.1A 2021-11-22 2021-11-22 Power grid load abnormity prediction method, system and storage medium Pending CN114091750A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111383136.1A CN114091750A (en) 2021-11-22 2021-11-22 Power grid load abnormity prediction method, system and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111383136.1A CN114091750A (en) 2021-11-22 2021-11-22 Power grid load abnormity prediction method, system and storage medium

Publications (1)

Publication Number Publication Date
CN114091750A true CN114091750A (en) 2022-02-25

Family

ID=80302592

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111383136.1A Pending CN114091750A (en) 2021-11-22 2021-11-22 Power grid load abnormity prediction method, system and storage medium

Country Status (1)

Country Link
CN (1) CN114091750A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116362300A (en) * 2022-06-29 2023-06-30 国网河南省电力公司 Regional power grid abnormal disturbance quantity prediction method, device, medium and electronic equipment

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116362300A (en) * 2022-06-29 2023-06-30 国网河南省电力公司 Regional power grid abnormal disturbance quantity prediction method, device, medium and electronic equipment
CN116362300B (en) * 2022-06-29 2024-02-09 国网河南省电力公司 Regional power grid abnormal disturbance quantity prediction method, device, medium and electronic equipment

Similar Documents

Publication Publication Date Title
Saini et al. Artificial neural network based peak load forecasting using Levenberg–Marquardt and quasi-Newton methods
Li et al. Adaptive online monitoring of voltage stability margin via local regression
CN104934968A (en) Multi-agent based distribution network disaster responding recovery coordinate control method and multi-agent based distribution network disaster responding recovery coordinate control device
CN115291116B (en) Energy storage battery health state prediction method and device and intelligent terminal
CN109782124B (en) Main distribution integrated fault positioning method and system based on gradient descent algorithm
Fischl et al. Screening power system contingencies using a back-propagation trained multiperceptron
CN115642706A (en) Power distribution load monitoring system in power grid
CN114266301A (en) Intelligent power equipment fault prediction method based on graph convolution neural network
CN116307641B (en) Digital power plant-oriented resource collaborative scheduling management method and system
CN115733730A (en) Power grid fault detection method and device based on graph neural network
CN106227127A (en) Generating equipment intelligent monitoring and controlling device and monitoring method
CN114091750A (en) Power grid load abnormity prediction method, system and storage medium
CN106548284A (en) A kind of adaptive mode massing power grid security Alarm Assessment method towards operation regulation and control
CN113825165B (en) 5G slice network congestion early warning method and device based on time diagram network
CN116780509A (en) Power grid random scene generation method integrating discrete probability and CGAN
CN117557047A (en) Power distribution equipment operation and maintenance optimization method and system based on deep reinforcement learning
CN117236380A (en) Power system fault prediction method, system, electronic equipment and medium
Naik Power system contingency ranking using Newton Raphson load flow method and its prediction using soft computing techniques
Tiwary et al. Multi-dimensional ANN application for active power flow state classification on a utility system
CN114090561A (en) Power grid data cleaning method, system and storage medium
CN115660893A (en) Transformer substation bus load prediction method based on load characteristics
CN113128130B (en) Real-time monitoring method and device for judging stability of direct-current power distribution system
Mahto et al. MPGCN-OPF: A message passing graph convolution approach for optimal power flow for distribution network
CN114266370A (en) Method and system for generating fault handling plan of power grid equipment in typhoon meteorological environment on line and storage medium
Liu et al. Evaluation of hurricane impact on failure rate of transmission lines using fuzzy expert system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination