CN117390836A - BEMD-based GRU interval system inertia evaluation and prediction method - Google Patents

BEMD-based GRU interval system inertia evaluation and prediction method Download PDF

Info

Publication number
CN117390836A
CN117390836A CN202311277960.8A CN202311277960A CN117390836A CN 117390836 A CN117390836 A CN 117390836A CN 202311277960 A CN202311277960 A CN 202311277960A CN 117390836 A CN117390836 A CN 117390836A
Authority
CN
China
Prior art keywords
value
interval
gru
prediction
inertia
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
CN202311277960.8A
Other languages
Chinese (zh)
Inventor
李智诚
高统彤
邓超平
张伟骏
陈大玮
张抒凌
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.)
Electric Power Research Institute of State Grid Fujian Electric Power Co Ltd
State Grid Fujian Electric Power Co Ltd
Original Assignee
Electric Power Research Institute of State Grid Fujian Electric Power Co Ltd
State Grid Fujian Electric Power 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 Electric Power Research Institute of State Grid Fujian Electric Power Co Ltd, State Grid Fujian Electric Power Co Ltd filed Critical Electric Power Research Institute of State Grid Fujian Electric Power Co Ltd
Priority to CN202311277960.8A priority Critical patent/CN117390836A/en
Publication of CN117390836A publication Critical patent/CN117390836A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Power Engineering (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to a GRU interval system inertia evaluation and prediction method based on BEMD, which comprises the following steps: acquiring original data related to inertia of a target system through a PMU synchronous phasor measurement device; constructing the collected original data into a complex value sequence; decomposing the constructed complex value sequence by using a BEMD binary empirical mode decomposition method to obtain a plurality of two-dimensional mode components, and reconstructing the plurality of two-dimensional mode components into components with low complexity; optimizing parameters of the GRU model by utilizing a sparrow search algorithm, modeling and predicting upper and lower boundaries of each reconstructed component by taking the optimized GRU model as a single predictor, evaluating and analyzing inertia prediction results, obtaining and outputting final prediction interval values of inertia.

Description

BEMD-based GRU interval system inertia evaluation and prediction method
Technical Field
The invention relates to a GRU interval system inertia evaluation and prediction method based on BEMD, and belongs to the technical field of power system optimization operation.
Background
The interval value sequence of the system inertia consists of inertia peaks and valleys in a certain period, is a special type of time sequence, and contains more fluctuation information (such as uncertainty, variability and trend) compared with the point sequence. The time series of interval values of the system inertia can reduce the amount of random variation relative to the classical point series. Therefore, if the inertia interval prediction result can be provided, a decision maker can better grasp the variation fluctuation range of the inertia of the future power system, and can better know potential uncertainty and risk factors of the future power grid when planning the power generation and analyzing system to be safe, so that a reasonable decision can be made in time.
The conventional inertia evaluation method is rarely concerned with the interval evaluation and prediction of the inertia of the system, and most of the conventional inertia evaluation methods do not consider possible correlation between upper and lower boundaries of the inertia interval. In order to improve the prediction precision and the model robustness of the inertia evaluation model, the possible relation between the upper limit and the lower limit of the interval is considered, and the invention provides the GRU interval system inertia evaluation and prediction method based on BEMD.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a GRU interval system inertia evaluation and prediction method based on BEMD, which can overcome the defect that the traditional method is difficult to process nonlinear and non-stable signals and time sequence relations in system inertia evaluation and prediction, and improves the prediction accuracy of the inertia interval range of an electric power system.
The technical scheme of the invention is as follows:
in one aspect, the invention provides a BEMD-based GRU interval system inertia evaluation and prediction method, which comprises the following steps:
acquiring original data related to inertia of a target system through a PMU synchronous phasor measurement device;
Constructing the collected original data into a complex value sequence;
decomposing the constructed complex value sequence by using a BEMD binary empirical mode decomposition method to obtain a plurality of two-dimensional mode components, and reconstructing the plurality of two-dimensional mode components into components with low complexity;
optimizing parameters of the GRU model by utilizing a sparrow search algorithm, modeling and predicting upper and lower boundaries of each reconstructed component by taking the optimized GRU model as a single predictor, evaluating and analyzing inertia prediction results, obtaining and outputting final prediction interval values of inertia.
As a preferred embodiment, the step of constructing the collected raw data into a complex value sequence specifically includes:
the method comprises the steps of carrying out missing value processing on collected original data, weighting sampling data of the same sampling point and sampling data of two adjacent sampling points at the previous moment aiming at missing sampling points in the original data, and taking the weighted sampling data as interpolation data, wherein the method specifically comprises the following steps:
wherein x (d, t) represents the interpolation data at the time of the day (d, t);
the collected original data is subjected to anomaly detection, the anomaly data is screened out and is modified by adopting a longitudinal comparison method, and the method specifically comprises the following steps:
wherein d represents a day, and n is an integer in the [2,5] interval;
Constructing interval time series by using a method for improving time scale, finding out the maximum value and the minimum value of each moment in the original data after the missing value processing and the abnormality detection as the upper limit and the lower limit of the interval variable, and generating the interval time series by arranging in time sequence at each time pointWherein->For the lower boundary of the interval, +.>For the upper boundary of the interval, +.>Converting the upper and lower bounds of the sequence into complex value sequences, respectivelyReal and imaginary parts of (a) are provided.
As a preferred embodiment, the method for decomposing the constructed complex-valued sequence by using the BEMD binary empirical mode decomposition method specifically includes:
performing BEMD decomposition on the complex value sequence, performing complexity analysis on the decomposed two-dimensional modal components by adopting a multi-element multi-scale permutation entropy method, adding the components with low complexity into a new component, reducing the calculation complexity, reducing the accumulation of prediction errors, and keeping the components with high complexity as a single component;
when the multi-element multi-scale arrangement entropy value is larger than or equal to the threshold value theta=0.5, the decomposed component is regarded as a component with high complexity;
The range of the scale factor s is defined as 1-10, and the components with low complexity are reconstructed, namely, each component with higher complexity in the binary modal components is used as an independent component, and the binary modal components with lower complexity are added into the new components.
In a preferred embodiment, the optimizing the parameters of the GRU model by using the sparrow search algorithm, and modeling and predicting the upper and lower bounds of each reconstructed component by using the optimized GRU model as a single predictor, and the step of evaluating and analyzing the inertia prediction result specifically includes:
the neuron number c of the first hidden layer of the GRU model 1 Number of neurons of second hidden layer c 2 Number of iterations c 3 And learning rate c 4 The four parameters are used as optimizing objects of an SSA sparrow searching algorithm;
taking the number of network parameters of the GRU model as the dimension component of the individual sparrow position, and taking the structural parameter C= (C) of the GRU 1 ,c 2 ,c 3 ,c 4 ) Initial position vector X encoded as individual sparrows ij ＝(X i1 ,X i2 ,X i3 ,X i4 )，X ij Indicating the j-th position of the i-th sparrow in four-dimensional space. Assume that a population consisting of n sparrows is:
wherein d is the dimension of the problem variable to be optimized, and n is the number of sparrows;
the fitness value f of all sparrows is:
During each iteration, the position of the finder is updated, and the formula is as follows:
wherein X is i,j For the ith sparrowAt the j-th dimension, t is the number of iterations, j=1, 2, 3.,. D, alpha is (0, 1)]A random number in between, M is the maximum iteration number, W 2 (W 2 ∈[0,1]) For the early warning value, ST (ST.epsilon.0.5, 1]) Q is a random number conforming to normal distribution, L is a 1 row and d column matrix, and each element value is 1;
initializing SSA parameters including sparrow population N, finder proportion PD, adder proportion 1-PD, early warning value ST and maximum iteration number M; taking four parameters to be optimized as the positions of sparrows, randomly generating position vectors of all sparrows within the range, determining an adaptation function of SSA, calculating an adaptation value of each initial sparrow, and sequencing the values, wherein the adaptation function of SSA takes an interval error rate of a load actual value and a predicted value of GRU as an adaptation function;
updating the positions of discoverers, joiners and scouters in the sparrows, calculating fitness values, finding out the optimal fitness values under the current iteration times in the population and the positions of the corresponding sparrows, judging whether the maximum iteration times are reached, stopping iteration if the conditions are met, outputting an optimal result, returning to the previous step to continue iteration optimizing if the conditions are not met, and updating the positions of the scouters in each iteration process as follows:
Wherein t is the iteration number, X best X is the globally optimal position worst Is the global worst position, k is [ -1,1]A random number in between, beta is a step control parameter, the value of which is a normal distribution random number obeying a mean value of 0 and a variance of 1, f g For the fitness value of the current global optimal position, f w For the fitness value of the current global worst position, f i The fitness value of the current sparrow individual position is epsilon which is a constant and is used for avoiding that the denominator is 0;
decoding the optimal solution of the SSA to obtain optimal parameters of the GRU model, and finally predicting the interval load data by using the GRU;
to verify the effect of the proposed algorithm on the interval prediction, an interval-type average relative error ARV will be employed I And a interval-type U value U of the Theil statistic I The two indexes evaluate and analyze the prediction result:
wherein m is the number of intervals in the test set,for the original interval of time t, +.>For the corresponding prediction value, +.>Mean value of samples, +.>And->The upper-and lower-bound averages are shown, respectively.
Wherein ARV I Is interval type average relative error, the lower the value is, the higher the prediction accuracy is; u (U) I For the interval-type U value of the Theil statistic, if the predictive performance of the algorithm model is better than the result of random walk, then U I Less than 1, if U I Greater than 1, the worse the model prediction performance, the closer its value is to 0, indicating higher prediction accuracy.
On the other hand, the invention also provides a GRU interval system inertia evaluation and prediction system based on BEMD, which comprises the following steps:
the acquisition module is used for acquiring the original data related to the inertia of the target system through the PMU synchronous phasor measurement device;
the sequence construction module is used for constructing the acquired original data into a complex value sequence;
the decomposition module is used for decomposing the constructed complex value sequence by using a BEMD binary empirical mode decomposition method to obtain a plurality of two-dimensional mode components, and reconstructing the plurality of two-dimensional mode components into components with low complexity;
and the evaluation prediction module optimizes parameters of the GRU model by utilizing a sparrow search algorithm, models and predicts upper and lower boundaries of each reconstructed component by taking the optimized GRU model as a single predictor, evaluates and analyzes an inertia prediction result, and obtains and outputs a final prediction interval value of inertia.
As a preferred embodiment, the step of the sequence construction module constructing the collected raw data into a complex value sequence specifically includes:
the method comprises the steps of carrying out missing value processing on collected original data, weighting sampling data of the same sampling point and sampling data of two adjacent sampling points at the previous moment aiming at missing sampling points in the original data, and taking the weighted sampling data as interpolation data, wherein the method specifically comprises the following steps:
Wherein x (d, t) represents the interpolation data at the time of the day (d, t);
the collected original data is subjected to anomaly detection, the anomaly data is screened out and is modified by adopting a longitudinal comparison method, and the method specifically comprises the following steps:
wherein d represents a day, and n is an integer in the [2,5] interval;
constructing interval time sequence by using method of increasing time scale, finding out missing value processing and abnormality detectionThe maximum value and the minimum value of each moment in the measured original data are used as the upper limit and the lower limit of the interval type variable, and at each time point, the interval type time series is generated by chronologically arrangingWherein->For the lower boundary of the interval, +.>For the upper boundary of the interval, +.>Converting the upper and lower bounds of the sequence into complex value sequences, respectivelyReal and imaginary parts of (a) are provided.
As a preferred embodiment, the method for decomposing the constructed complex-valued sequence by the decomposition module by using the BEMD binary empirical mode decomposition method specifically includes:
performing BEMD decomposition on the complex value sequence, performing complexity analysis on the decomposed two-dimensional modal components by adopting a multi-element multi-scale permutation entropy method, adding the components with low complexity into a new component, reducing the calculation complexity, reducing the accumulation of prediction errors, and keeping the components with high complexity as a single component;
When the multi-element multi-scale arrangement entropy value is larger than or equal to the threshold value theta=0.5, the decomposed component is regarded as a component with high complexity;
the range of the scale factor s is defined as 1-10, and the components with low complexity are reconstructed, namely, each component with higher complexity in the binary modal components is used as an independent component, and the binary modal components with lower complexity are added into the new components.
In a preferred embodiment, the evaluation prediction module optimizes parameters of the GRU model by using a sparrow search algorithm, and uses the optimized GRU model as a single predictor to perform modeling prediction on upper and lower bounds of each reconstructed component, and the step of performing evaluation analysis on inertia prediction results specifically includes:
the neuron number c of the first hidden layer of the GRU model 1 Number of neurons of second hidden layer c 2 Number of iterations c 3 And learning rate c 4 The four parameters are used as optimizing objects of an SSA sparrow searching algorithm;
taking the number of network parameters of the GRU model as the dimension component of the individual sparrow position, and taking the structural parameter C= (C) of the GRU 1 ,c 2 ,c 3 ,c 4 ) Initial position vector X encoded as individual sparrows ij ＝(X i1 ,X i2 ,X i3 ,X i4 )，X ij Indicating the j-th position of the i-th sparrow in four-dimensional space. Assume that a population consisting of n sparrows is:
wherein d is the dimension of the problem variable to be optimized, and n is the number of sparrows;
the fitness value f of all sparrows is:
during each iteration, the position of the finder is updated, and the formula is as follows:
wherein X is i,j For the position of the ith sparrow in the j-th dimension, t is the number of iterations, j=1, 2,3, d, a is (0, 1]A random number in between, M is the maximum iterationTimes, W 2 (W 2 ∈[0,1]) For the early warning value, ST (ST.epsilon.0.5, 1]) Q is a random number conforming to normal distribution, L is a 1 row and d column matrix, and each element value is 1;
initializing SSA parameters including sparrow population N, finder proportion PD, adder proportion 1-PD, early warning value ST and maximum iteration number M; taking four parameters to be optimized as the positions of sparrows, randomly generating position vectors of all sparrows within the range, determining an adaptation function of SSA, calculating an adaptation value of each initial sparrow, and sequencing the values, wherein the adaptation function of SSA takes an interval error rate of a load actual value and a predicted value of GRU as an adaptation function;
Updating the positions of discoverers, joiners and scouters in the sparrows, calculating fitness values, finding out the optimal fitness values under the current iteration times in the population and the positions of the corresponding sparrows, judging whether the maximum iteration times are reached, stopping iteration if the conditions are met, outputting an optimal result, returning to the previous step to continue iteration optimizing if the conditions are not met, and updating the positions of the scouters in each iteration process as follows:
wherein t is the iteration number, X best X is the globally optimal position worst Is the global worst position, k is [ -1,1]A random number in between, beta is a step control parameter, the value of which is a normal distribution random number obeying a mean value of 0 and a variance of 1, f g For the fitness value of the current global optimal position, f w For the fitness value of the current global worst position, f i The fitness value of the current sparrow individual position is epsilon which is a constant and is used for avoiding that the denominator is 0;
decoding the optimal solution of the SSA to obtain optimal parameters of the GRU model, and finally predicting the interval load data by using the GRU;
to verify the effect of the proposed algorithm on interval prediction, interval-type average relative errors will be employedDifference ARV I And a interval-type U value U of the Theil statistic I The two indexes evaluate and analyze the prediction result:
wherein m is the number of intervals in the test set,for the original interval of time t, +.>For the corresponding prediction value, +.>Mean value of samples, +.>And->The upper-and lower-bound averages are shown, respectively.
Wherein ARV I Is interval type average relative error, the lower the value is, the higher the prediction accuracy is; u (U) I For the interval-type U value of the Theil statistic, if the predictive performance of the algorithm model is better than the result of random walk, then U I Less than 1, if U I Greater than 1, the worse the model prediction performance, the closer its value is to 0, indicating higher prediction accuracy.
In still another aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor implements the BEMD-based GRU interval system inertia evaluation and prediction method according to any of the embodiments of the present invention when the program is executed by the processor.
In yet another aspect, the present invention further provides a computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a method for evaluating and predicting the inertia of a stu interval system based on BEMD according to any embodiment of the present invention.
The invention has the following beneficial effects:
by applying the BEMD technology, the invention can effectively extract the local characteristics of the inertia of the system, thereby improving the understanding and modeling capability of the dynamic characteristics of the system. And secondly, a GRU neural network structure is introduced, so that the time sequence change and the nonlinear relation of the system inertia can be captured, the accurate evaluation and prediction of the system inertia are realized, and the stable operation and the optimal scheduling of the power system are facilitated. In addition, the method has stronger robustness, can process the influence of noise and incomplete data, improves the reliability of evaluation and prediction, can help a system decision maker to know the inertia level of the system in real time, and provides important support for the maintenance of the frequency stability of the system.
Drawings
FIG. 1 is a schematic flow chart of a method according to a first embodiment of the invention;
FIG. 2 is a flow chart of optimizing GRU model and predicting GRU model by the optimized GRU model.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the step numbers used herein are for convenience of description only and are not limiting as to the order in which the steps are performed.
It is to be understood that the terminology used in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
The terms "comprises" and "comprising" indicate the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The term "and/or" refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Embodiment one:
referring to fig. 1, a BEMD-based method for evaluating and predicting inertia of a GRU interval system includes the following steps:
s100, acquiring original data related to inertia of a target system through a PMU synchronous phasor measurement device;
S200, constructing the acquired original data into a complex value sequence;
s300, decomposing the constructed complex value sequence by using a BEMD binary empirical mode decomposition method to obtain a plurality of two-dimensional mode components, and reconstructing the plurality of two-dimensional mode components into components with low complexity;
and S400, optimizing parameters of the GRU model by utilizing a sparrow search algorithm, modeling and predicting the upper and lower boundaries of each reconstructed component by taking the optimized GRU model as a single predictor, evaluating and analyzing an inertia prediction result, obtaining a final prediction interval value of inertia, and outputting the final prediction interval value of inertia.
As a preferred embodiment, the step of constructing the collected raw data into a complex value sequence specifically includes:
the method comprises the steps of carrying out missing value processing on collected original data, weighting sampling data of the same sampling point and sampling data of two adjacent sampling points at the previous moment aiming at missing sampling points in the original data, and taking the weighted sampling data as interpolation data, wherein the method specifically comprises the following steps:
wherein x (d, t) represents the interpolation data at the time of the day (d, t);
the collected original data is subjected to anomaly detection, the anomaly data is screened out and is modified by adopting a longitudinal comparison method, and the method specifically comprises the following steps:
wherein d represents a day, and n is an integer in the [2,5] interval;
Constructing interval time series by using a method for improving time scale, finding out the maximum value and the minimum value of each moment in the original data after the missing value processing and the abnormality detection as the upper limit and the lower limit of the interval variable, and generating the interval time series by arranging in time sequence at each time pointWherein->For the lower boundary of the interval, +.>For the upper boundary of the interval, +.>Converting the upper and lower bounds of the sequence into complex value sequences, respectivelyReal and imaginary parts of (a) are provided.
As a preferred embodiment, the method for decomposing the constructed complex-valued sequence by using the BEMD binary empirical mode decomposition method specifically includes:
performing BEMD decomposition on the complex value sequence, performing complexity analysis on the decomposed two-dimensional modal components by adopting a multi-element multi-scale permutation entropy method, adding the components with low complexity into a new component, reducing the calculation complexity, reducing the accumulation of prediction errors, and keeping the components with high complexity as a single component;
when the multi-element multi-scale arrangement entropy value is larger than or equal to the threshold value theta=0.5, the decomposed component is regarded as a component with high complexity;
The range of the scale factor s is defined as 1-10, and the components with low complexity are reconstructed, namely, each component with higher complexity in the binary modal components is used as an independent component, and the binary modal components with lower complexity are added into the new components. The components with lower complexity are reconstructed based on the steps, so that the calculation complexity and estimation error accumulation can be reduced.
Referring specifically to fig. 2, as a preferred implementation manner of this embodiment, the steps of optimizing parameters of the GRU model by using a sparrow search algorithm, modeling and predicting upper and lower bounds of each reconstructed component by using the optimized GRU model as a single predictor, and performing evaluation and analysis on inertia prediction results are specifically as follows:
the neuron number c of the first hidden layer of the GRU model 1 Number of neurons of second hidden layer c 2 Number of iterations c 3 And learning rate c 4 The four parameters are used as optimizing objects of an SSA sparrow searching algorithm;
taking the number of network parameters of the GRU model as the dimension component of the individual sparrow position, and taking the structural parameter C= (C) of the GRU 1 ,c 2 ,c 3 ,c 4 ) Initial position vector X encoded as individual sparrows ij ＝(X i1 ,X i2 ,X i3 ,X i4 )，X ij Indicating the j-th position of the i-th sparrow in four-dimensional space. Assume that a population consisting of n sparrows is:
Wherein d is the dimension of the problem variable to be optimized, and n is the number of sparrows;
the fitness value f of all sparrows is:
during each iteration, the position of the finder is updated, and the formula is as follows:
wherein X is i,j For the position of the ith sparrow in the j-th dimension, t is the number of iterations, j=1, 2,3, d, a is (0, 1]A random number in between, M is the maximum iteration number, W 2 (W 2 ∈[0,1]) For the early warning value, ST (ST.epsilon.0.5, 1]) Q is a random number conforming to normal distribution, L is a 1 row and d column matrix, and each element value is 1;
initializing SSA parameters including sparrow population N, finder proportion PD, adder proportion 1-PD, early warning value ST and maximum iteration number M; taking four parameters to be optimized as the positions of sparrows, randomly generating position vectors of all sparrows within the range, determining an fitness function of SSA, calculating an initial fitness value of each sparrow, sequencing the values, wherein the inertia value is influenced by a load value, and generally the larger the load is, the larger the required inertia is, so that the embodiment takes the interval error rate of a load predicted value and a load actual value of GRU, namely the deviation as the fitness function;
Updating the positions of discoverers, joiners and scouters in the sparrows, calculating fitness values, finding out the optimal fitness values under the current iteration times in the population and the positions of the corresponding sparrows, judging whether the maximum iteration times are reached, stopping iteration if the conditions are met, outputting an optimal result, returning to the previous step to continue iteration optimizing if the conditions are not met, and updating the positions of the scouters in each iteration process as follows:
wherein t is the iteration number, X best X is the globally optimal position worst Is the global worst position, k is [ -1,1]A random number in between, beta is a step control parameter, the value of which is a normal distribution random number obeying a mean value of 0 and a variance of 1, f g For the fitness value of the current global optimal position, f w For the fitness value of the current global worst position, f i The fitness value of the current sparrow individual position is epsilon which is a constant and is used for avoiding that the denominator is 0;
decoding the optimal solution of the SSA to obtain optimal parameters of the GRU model, and finally predicting the interval load data by using the GRU;
to verify the effect of the proposed algorithm on the interval prediction, an interval-type average relative error ARV will be employed I And a interval-type U value U of the Theil statistic I The two indexes evaluate and analyze the prediction result:
wherein m is the number of intervals in the test set,for the original interval of time t, +.>For the corresponding prediction value, +.>Mean value of samples, +.>And->The upper-and lower-bound averages are shown, respectively.
Wherein ARV I Is interval type average relative error, the lower the value is, the higher the prediction accuracy is; u (U) I For the interval-type U value of the Theil statistic, if the predictive performance of the algorithm model is better than the result of random walk, then U I Less than 1, if U I Greater than 1, the worse the model prediction performance, the closer its value is to 0, indicating higher prediction accuracy.
Embodiment two:
the present embodiment provides a BEMD-based inertia evaluation and prediction system for a GRU interval system, including:
the acquisition module is used for acquiring the original data related to the inertia of the target system through the PMU synchronous phasor measurement device; the module is used for implementing the function of step S100 in the first embodiment, and will not be described here again;
the sequence construction module is used for constructing the acquired original data into a complex value sequence; the module is used for implementing the function of step S200 in the first embodiment, and will not be described in detail herein;
the decomposition module is used for decomposing the constructed complex value sequence by using a BEMD binary empirical mode decomposition method to obtain a plurality of two-dimensional mode components, and reconstructing the plurality of two-dimensional mode components into components with low complexity; the module is used for implementing the function of step S300 in the first embodiment, and will not be described in detail herein;
The assessment prediction module optimizes parameters of the GRU model by utilizing a sparrow search algorithm, models and predicts upper and lower boundaries of each reconstructed component by taking the optimized GRU model as a single predictor, and carries out assessment analysis on inertia prediction results to obtain and output final prediction interval values of inertia; this module is used to implement the function of step S400 in the first embodiment, and will not be described here again.
As a preferred implementation manner of this embodiment, the step of the sequence construction module constructing the collected raw data into a complex value sequence specifically includes:
the method comprises the steps of carrying out missing value processing on collected original data, weighting sampling data of the same sampling point and sampling data of two adjacent sampling points at the previous moment aiming at missing sampling points in the original data, and taking the weighted sampling data as interpolation data, wherein the method specifically comprises the following steps:
wherein x (d, t) represents the interpolation data at the time of the day (d, t);
the collected original data is subjected to anomaly detection, the anomaly data is screened out and is modified by adopting a longitudinal comparison method, and the method specifically comprises the following steps:
wherein d represents a day, and n is an integer in the [2,5] interval;
constructing interval time series by using a method for improving time scale, finding out the maximum value and the minimum value of each moment in the original data after the missing value processing and the abnormality detection as the upper limit and the lower limit of the interval variable, and generating the interval time series by arranging in time sequence at each time point Wherein->For the lower boundary of the interval, +.>For the upper boundary of the interval, +.>Converting the upper and lower bounds of the sequence into complex value sequences, respectivelyReal and imaginary parts of (a) are provided.
As a preferred implementation manner of this embodiment, the method for decomposing the constructed complex-valued sequence by the decomposition module using the BEMD binary empirical mode decomposition method specifically includes:
performing BEMD decomposition on the complex value sequence, performing complexity analysis on the decomposed two-dimensional modal components by adopting a multi-element multi-scale permutation entropy method, adding the components with low complexity into a new component, reducing the calculation complexity, reducing the accumulation of prediction errors, and keeping the components with high complexity as a single component;
when the multi-element multi-scale arrangement entropy value is larger than or equal to the threshold value theta=0.5, the decomposed component is regarded as a component with high complexity;
the range of the scale factor s is defined as 1-10, and the components with low complexity are reconstructed, namely, each component with higher complexity in the binary modal components is used as an independent component, and the binary modal components with lower complexity are added into the new components.
As a preferred implementation manner of this embodiment, the evaluation prediction module optimizes parameters of the GRU model by using a sparrow search algorithm, and uses the optimized GRU model as a single predictor to perform modeling prediction on upper and lower bounds of each reconstructed component, and the step of performing evaluation analysis on an inertia prediction result specifically includes:
the neuron number c of the first hidden layer of the GRU model 1 Number of neurons of second hidden layer c 2 Number of iterations c 3 And learning rate c 4 The four parameters are used as optimizing objects of an SSA sparrow searching algorithm;
the number of network parameters of the GRU model is used as the dimension component of the sparrow individual position,the structural parameter c= (C) of the GRU 1 ,c 2 ,c 3 ,c 4 ) Initial position vector X encoded as individual sparrows ij ＝(X i1 ,X i2 ,X i3 ,X i4 )，X ij Indicating the j-th position of the i-th sparrow in four-dimensional space. Assume that a population consisting of n sparrows is:
wherein d is the dimension of the problem variable to be optimized, and n is the number of sparrows;
the fitness value f of all sparrows is:
during each iteration, the position of the finder is updated, and the formula is as follows:
wherein X is i,j For the position of the ith sparrow in the j-th dimension, t is the number of iterations, j=1, 2,3, d, a is (0, 1]A random number in between, M is the maximum iteration number, W 2 (W 2 ∈[0,1]) For the early warning value, ST (ST.epsilon.0.5, 1]) Q is a random number conforming to normal distribution, L is a 1 row and d column matrix, and each element value is 1;
initializing SSA parameters including sparrow population N, finder proportion PD, adder proportion 1-PD, early warning value ST and maximum iteration number M; taking four parameters to be optimized as the positions of sparrows, randomly generating position vectors of all sparrows within the range, determining an adaptation function of SSA, calculating an adaptation value of each initial sparrow, and sequencing the values, wherein the adaptation function of SSA takes an interval error rate of a load actual value and a predicted value of GRU as an adaptation function;
updating the positions of discoverers, joiners and scouters in the sparrows, calculating fitness values, finding out the optimal fitness values under the current iteration times in the population and the positions of the corresponding sparrows, judging whether the maximum iteration times are reached, stopping iteration if the conditions are met, outputting an optimal result, returning to the previous step to continue iteration optimizing if the conditions are not met, and updating the positions of the scouters in each iteration process as follows:
wherein t is the iteration number, X best X is the globally optimal position worst Is the global worst position, k is [ -1,1]A random number in between, beta is a step control parameter, the value of which is a normal distribution random number obeying a mean value of 0 and a variance of 1, f g For the fitness value of the current global optimal position, f w For the fitness value of the current global worst position, f i The fitness value of the current sparrow individual position is epsilon which is a constant and is used for avoiding that the denominator is 0;
decoding the optimal solution of the SSA to obtain optimal parameters of the GRU model, and finally predicting the interval load data by using the GRU;
to verify the effect of the proposed algorithm on the interval prediction, an interval-type average relative error ARV will be employed I And a interval-type U value U of the Theil statistic I The two indexes evaluate and analyze the prediction result:
wherein m is the number of intervals in the test set,for the original interval of time t, +.>For the corresponding prediction value, +.>Mean value of samples, +.>And->The upper-and lower-bound averages are shown, respectively.
Wherein ARV I Is interval type average relative error, the lower the value is, the higher the prediction accuracy is; u (U) I For the interval-type U value of the Theil statistic, if the predictive performance of the algorithm model is better than the result of random walk, then U I Less than 1, if U I Greater than 1, the worse the model prediction performance, the closer its value is to 0, indicating higher prediction accuracy.
Embodiment III:
the present embodiment proposes an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, said processor implementing a method according to any of the embodiments of the invention when executing said program.
Embodiment four:
the present embodiment proposes a computer readable storage medium on which a computer program is stored, which when executed by a processor implements a method according to any of the embodiments of the invention.
In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "and/or", describes an association relation of association objects, and indicates that there may be three kinds of relations, for example, a and/or B, and may indicate that a alone exists, a and B together, and B alone exists. Wherein A, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of the following" and the like means any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c may represent: a, b, c, a and b, a and c, b and c or a and b and c, wherein a, b and c can be single or multiple.
Those of ordinary skill in the art will appreciate that the various elements and algorithm steps described in the embodiments disclosed herein can be implemented as a combination of electronic hardware, computer software, and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In several embodiments provided herein, any of the functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (hereinafter referred to as ROM), a random access Memory (Random Access Memory) and various media capable of storing program codes such as a magnetic disk or an optical disk.
The foregoing description is only illustrative of the present invention and is not intended to limit the scope of the invention, and all equivalent structures or equivalent processes or direct or indirect application in other related technical fields are included in the scope of the present invention.

Claims (10)

1. The GRU interval system inertia evaluation and prediction method based on BEMD is characterized by comprising the following steps:
acquiring original data related to inertia of a target system through a PMU synchronous phasor measurement device;
constructing the collected original data into a complex value sequence;
decomposing the constructed complex value sequence by using a BEMD binary empirical mode decomposition method to obtain a plurality of two-dimensional mode components, and reconstructing the plurality of two-dimensional mode components into components with low complexity;
optimizing parameters of the GRU model by utilizing a sparrow search algorithm, modeling and predicting upper and lower boundaries of each reconstructed component by taking the optimized GRU model as a single predictor, evaluating and analyzing inertia prediction results, obtaining and outputting final prediction interval values of inertia.
2. The method for evaluating and predicting the inertia of a cru interval system based on BEMD as claimed in claim 1, wherein said step of constructing the collected raw data into a complex-valued sequence is specifically as follows:
The method comprises the steps of carrying out missing value processing on collected original data, weighting sampling data of the same sampling point and sampling data of two adjacent sampling points at the previous moment aiming at missing sampling points in the original data, and taking the weighted sampling data as interpolation data, wherein the method specifically comprises the following steps:
wherein x (d, t) represents the interpolation data at the time of the day (d, t);
the collected original data is subjected to anomaly detection, the anomaly data is screened out and is modified by adopting a longitudinal comparison method, and the method specifically comprises the following steps:
wherein d represents a day, and n is an integer in the [2,5] interval;
constructing interval time series by using a method for improving time scale, finding out the maximum value and the minimum value of each moment in the original data after the missing value processing and the abnormality detection as the upper limit and the lower limit of the interval variable, and generating the interval time series by arranging in time sequence at each time pointt=1, 2..n, wherein +.>For the lower boundary of the interval, +.>For the upper boundary of the interval, +.>Converting the upper and lower bounds of the sequence into complex value sequences, respectivelyt=1, 2..real and imaginary parts of n.
3. The method for evaluating and predicting the inertia of the GRU interval system based on the BEMD according to claim 1, wherein the method for decomposing the constructed complex-valued sequence by using the BEMD binary empirical mode decomposition method specifically comprises the following steps:
Performing BEMD decomposition on the complex value sequence, performing complexity analysis on the decomposed two-dimensional modal components by adopting a multi-element multi-scale permutation entropy method, adding the components with low complexity into a new component, reducing the calculation complexity, reducing the accumulation of prediction errors, and keeping the components with high complexity as a single component;
when the multi-element multi-scale arrangement entropy value is larger than or equal to the threshold value theta=0.5, the decomposed component is regarded as a component with high complexity;
the range of the scale factor s is defined as 1-10, and the components with low complexity are reconstructed, namely, each component with higher complexity in the binary modal components is used as an independent component, and the binary modal components with lower complexity are added into the new components.
4. The method for evaluating and predicting the inertia of the stump-based GRU section system according to claim 1, wherein the optimizing the parameters of the GRU model by using the sparrow search algorithm and modeling and predicting the upper and lower bounds of each reconstructed component by using the optimized GRU model as a single predictor, and the evaluating and analyzing the inertia prediction result comprises the following steps:
The neuron number c of the first hidden layer of the GRU model 1 Number of neurons of second hidden layer c 2 Number of iterations c 3 And learning rate c 4 The four parameters are used as optimizing objects of an SSA sparrow searching algorithm;
taking the number of network parameters of the GRU model as the dimension component of the individual sparrow position, and taking the structural parameter C= (C) of the GRU 1 ,c 2 ,c 3 ,c 4 ) Initial position vector X encoded as individual sparrows ij ＝(X i1 ,X i2 ,X i3 ,X i4 )，X ij Indicating the j-th position of the i-th sparrow in four-dimensional space. Assume that a population consisting of n sparrows is:
wherein d is the dimension of the problem variable to be optimized, and n is the number of sparrows;
the fitness value f of all sparrows is:
during each iteration, the position of the finder is updated, and the formula is as follows:
wherein X is i,j For the position of the ith sparrow in the j-th dimension, t is the number of iterations, j=1, 2,3, d, a is (0, 1]A random number in between, M is the maximum iteration number, W 2 (W 2 ∈[0,1]) For the early warning value, ST (ST.epsilon.0.5, 1]) Q is a random number conforming to normal distribution, L is a 1 row and d column matrix, and each element value is 1;
initializing SSA parameters including sparrow population N, finder proportion PD, adder proportion 1-PD, early warning value ST and maximum iteration number M; taking four parameters to be optimized as the positions of sparrows, randomly generating position vectors of all sparrows within the range, determining an adaptation function of SSA, calculating an adaptation value of each initial sparrow, and sequencing the values, wherein the adaptation function of SSA takes an interval error rate of a load actual value and a predicted value of GRU as an adaptation function;
Updating the positions of discoverers, joiners and scouters in the sparrows, calculating fitness values, finding out the optimal fitness values under the current iteration times in the population and the positions of the corresponding sparrows, judging whether the maximum iteration times are reached, stopping iteration if the conditions are met, outputting an optimal result, returning to the previous step to continue iteration optimizing if the conditions are not met, and updating the positions of the scouters in each iteration process as follows:
wherein t is the iteration number, X best X is the globally optimal position worst Is the global worst position, k is [ -1,1]A random number in between, beta is a step control parameter, the value of which is a normal distribution random number obeying a mean value of 0 and a variance of 1, f g For the fitness value of the current global optimal position, f w For the fitness value of the current global worst position, f i The fitness value of the current sparrow individual position is epsilon which is a constant and is used for avoiding that the denominator is 0;
decoding the optimal solution of the SSA to obtain optimal parameters of the GRU model, and finally predicting the interval load data by using the GRU;
to verify the effect of the proposed algorithm on the interval prediction, an interval-type average relative error ARV will be employed I And a interval-type U value U of the Theil statistic I The two indexes evaluate and analyze the prediction result:
wherein m is the number of intervals in the test set,for the original interval of time t, +.>For the corresponding prediction value, +.>Mean value of samples, +.>And->The upper-and lower-bound averages are shown, respectively.
Wherein ARV I Is interval type average relative error, the lower the value is, the higher the prediction accuracy is; u (U) I For the interval-type U value of the Theil statistic, if the predictive performance of the algorithm model is better than the result of random walk, then U I Less than 1, if U I Greater than 1, the worse the model prediction performance, the closer its value is to 0, indicating higher prediction accuracy.
5. A BEMD-based stu interval system inertia evaluation and prediction system, comprising:
the acquisition module is used for acquiring the original data related to the inertia of the target system through the PMU synchronous phasor measurement device;
the sequence construction module is used for constructing the acquired original data into a complex value sequence;
the decomposition module is used for decomposing the constructed complex value sequence by using a BEMD binary empirical mode decomposition method to obtain a plurality of two-dimensional mode components, and reconstructing the plurality of two-dimensional mode components into components with low complexity;
and the evaluation prediction module optimizes parameters of the GRU model by utilizing a sparrow search algorithm, models and predicts upper and lower boundaries of each reconstructed component by taking the optimized GRU model as a single predictor, evaluates and analyzes an inertia prediction result, and obtains and outputs a final prediction interval value of inertia.
6. The BEMD-based GRU interval system inertia assessment and prediction system of claim 5, wherein said sequence construction module constructs the collected raw data into a complex sequence of values specifically as follows:
the method comprises the steps of carrying out missing value processing on collected original data, weighting sampling data of the same sampling point and sampling data of two adjacent sampling points at the previous moment aiming at missing sampling points in the original data, and taking the weighted sampling data as interpolation data, wherein the method specifically comprises the following steps:
wherein x (d, t) represents the interpolation data at the time of the day (d, t);
the collected original data is subjected to anomaly detection, the anomaly data is screened out and is modified by adopting a longitudinal comparison method, and the method specifically comprises the following steps:
wherein d represents a day, and n is an integer in the [2,5] interval;
constructing interval time series by using a method for improving time scale, finding out the maximum value and the minimum value of each moment in the original data after the missing value processing and the abnormality detection as the upper limit and the lower limit of the interval variable, and generating the interval time series by arranging in time sequence at each time pointt=1, 2..n, wherein +.>For the lower boundary of the interval, +.>For the upper boundary of the interval, +.>Converting the upper and lower bounds of the sequence into complex value sequences, respectively t=1, 2..real and imaginary parts of n.
7. The BEMD-based system inertia evaluation and prediction system of a GRU interval system of claim 5, wherein the decomposing module decomposes the constructed complex-valued sequence by using BEMD binary empirical mode decomposition method specifically comprises:
performing BEMD decomposition on the complex value sequence, performing complexity analysis on the decomposed two-dimensional modal components by adopting a multi-element multi-scale permutation entropy method, adding the components with low complexity into a new component, reducing the calculation complexity, reducing the accumulation of prediction errors, and keeping the components with high complexity as a single component;
when the multi-element multi-scale arrangement entropy value is larger than or equal to the threshold value theta=0.5, the decomposed component is regarded as a component with high complexity;
the range of the scale factor s is defined as 1-10, and the components with low complexity are reconstructed, namely, each component with higher complexity in the binary modal components is used as an independent component, and the binary modal components with lower complexity are added into the new components.
8. The BEMD-based system for evaluating and predicting inertia of a GRU interval system of claim 5, wherein said evaluating and predicting module optimizes parameters of a GRU model by using a sparrow search algorithm, and uses the optimized GRU model as a single predictor to model and predict upper and lower bounds of each reconstructed component, and the step of evaluating and analyzing inertia prediction results is specifically as follows:
the neuron number c of the first hidden layer of the GRU model 1 Number of neurons of second hidden layer c 2 Number of iterations c 3 And learning rate c 4 The four parameters are used as optimizing objects of an SSA sparrow searching algorithm;
network parameters of GRU modelThe number is taken as the dimension component of the individual sparrow position, and the structural parameter C= (C) of the GRU 1 ,c 2 ,c 3 ,c 4 ) Initial position vector X encoded as individual sparrows ij ＝(X i1 ,X i2 ,X i3 ,X i4 )，X ij Indicating the j-th position of the i-th sparrow in four-dimensional space. Assume that a population consisting of n sparrows is:
wherein d is the dimension of the problem variable to be optimized, and n is the number of sparrows;
the fitness value f of all sparrows is:
during each iteration, the position of the finder is updated, and the formula is as follows:
wherein X is i,j For the position of the ith sparrow in the j-th dimension, t is the number of iterations, j=1, 2,3, d, a is (0, 1 ]A random number in between, M is the maximum iteration number, W 2 (W 2 ∈[0,1]) For the early warning value, ST (ST.epsilon.0.5, 1]) Q is a random number conforming to normal distribution, L is a 1 row and d column matrix, and each element value is 1;
initializing SSA parameters including sparrow population N, finder proportion PD, adder proportion 1-PD, early warning value ST and maximum iteration number M; taking four parameters to be optimized as the positions of sparrows, randomly generating position vectors of all sparrows within the range, determining an adaptation function of SSA, calculating an adaptation value of each initial sparrow, and sequencing the values, wherein the adaptation function of SSA takes an interval error rate of a load actual value and a predicted value of GRU as an adaptation function;
updating the positions of discoverers, joiners and scouters in the sparrows, calculating fitness values, finding out the optimal fitness values under the current iteration times in the population and the positions of the corresponding sparrows, judging whether the maximum iteration times are reached, stopping iteration if the conditions are met, outputting an optimal result, returning to the previous step to continue iteration optimizing if the conditions are not met, and updating the positions of the scouters in each iteration process as follows:
Wherein t is the iteration number, X best X is the globally optimal position worst Is the global worst position, k is [ -1,1]A random number in between, beta is a step control parameter, the value of which is a normal distribution random number obeying a mean value of 0 and a variance of 1, f g For the fitness value of the current global optimal position, f w For the fitness value of the current global worst position, f i The fitness value of the current sparrow individual position is epsilon which is a constant and is used for avoiding that the denominator is 0;
decoding the optimal solution of the SSA to obtain optimal parameters of the GRU model, and finally predicting the interval load data by using the GRU;
to verify the effect of the proposed algorithm on the interval prediction, an interval-type average relative error ARV will be employed I And a interval-type U value U of the Theil statistic I The two indexes evaluate and analyze the prediction result:
wherein m is the set of testsThe number of the intervals is set,for the original interval of time t, +.>For the corresponding prediction value, +.>Mean value of samples, +.>And->The upper-and lower-bound averages are shown, respectively.
Wherein ARV I Is interval type average relative error, the lower the value is, the higher the prediction accuracy is; u (U) I For the interval-type U value of the Theil statistic, if the predictive performance of the algorithm model is better than the result of random walk, then U I Less than 1, if U I Greater than 1, the worse the model prediction performance, the closer its value is to 0, indicating higher prediction accuracy.
9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the BEMD-based GRU interval system inertia evaluation and prediction method of any one of claims 1 to 4 when the program is executed by the processor.
10. A computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the BEMD based GRU interval system inertia evaluation and prediction method of any one of claims 1 to 4.
CN202311277960.8A 2023-09-28 2023-09-28 BEMD-based GRU interval system inertia evaluation and prediction method Pending CN117390836A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311277960.8A CN117390836A (en) 2023-09-28 2023-09-28 BEMD-based GRU interval system inertia evaluation and prediction method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311277960.8A CN117390836A (en) 2023-09-28 2023-09-28 BEMD-based GRU interval system inertia evaluation and prediction method

Publications (1)

Publication Number Publication Date
CN117390836A true CN117390836A (en) 2024-01-12

Family

ID=89462216

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311277960.8A Pending CN117390836A (en) 2023-09-28 2023-09-28 BEMD-based GRU interval system inertia evaluation and prediction method

Country Status (1)

Country Link
CN (1) CN117390836A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117972433A (en) * 2024-03-28 2024-05-03 北京市农林科学院智能装备技术研究中心 Training method of mushroom room temperature prediction model, mushroom room temperature prediction method and device

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117972433A (en) * 2024-03-28 2024-05-03 北京市农林科学院智能装备技术研究中心 Training method of mushroom room temperature prediction model, mushroom room temperature prediction method and device

Similar Documents

Publication Publication Date Title
Galelli et al. An evaluation framework for input variable selection algorithms for environmental data-driven models
CN110309603B (en) Short-term wind speed prediction method and system based on wind speed characteristics
CN113297801A (en) Marine environment element prediction method based on STEOF-LSTM
CN113723007A (en) Mechanical equipment residual life prediction method based on DRSN and sparrow search optimization BilSTM
CN114297036A (en) Data processing method and device, electronic equipment and readable storage medium
CN116346639A (en) Network traffic prediction method, system, medium, equipment and terminal
CN110458278A (en) Degradation prediction method based on quantum attention loop coding decoding neural network
CN115983087A (en) Method for detecting time sequence data abnormity by combining attention mechanism and LSTM and terminal
Ghosh et al. Strong selection consistency of Bayesian vector autoregressive models based on a pseudo-likelihood approach
CN115456245A (en) Prediction method for dissolved oxygen in tidal river network area
CN115544890A (en) Short-term power load prediction method and system
CN117791570A (en) Short-term power load prediction system and method
CN117458955A (en) Operation control method and system of motor
Shang et al. Fault detection and identification based on explicit polynomial mapping and combined statistic in nonlinear dynamic processes
CN117390836A (en) BEMD-based GRU interval system inertia evaluation and prediction method
CN115389743A (en) Method, medium and system for predicting content interval of dissolved gas in transformer oil
CN115326397A (en) Method for establishing crankshaft bearing wear degree prediction model and prediction method and related device
CN115600500A (en) Ultrashort-term probability wind power prediction method based on space-time multiscale and K-SDW
CN113011086B (en) Estimation method of forest biomass based on GA-SVR algorithm
Yang et al. A novel decentralized weighted ReliefF-PCA method for fault detection
CN115249059A (en) Model training and abnormal data analysis method and device and computer storage medium
CN116700213B (en) Industrial equipment abnormality detection method and related device based on gating circulation unit
CN117494573B (en) Wind speed prediction method and system and electronic equipment
CN117647725B (en) Aging test method and system for PCBA
CN118211719A (en) Industrial short-term power load interval prediction method, device, equipment and medium

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