CN116702609A - ARIMA-LSTM combination algorithm-based carbon fiber composite core wire sag prediction method - Google Patents
ARIMA-LSTM combination algorithm-based carbon fiber composite core wire sag prediction method Download PDFInfo
- Publication number
- CN116702609A CN116702609A CN202310672265.5A CN202310672265A CN116702609A CN 116702609 A CN116702609 A CN 116702609A CN 202310672265 A CN202310672265 A CN 202310672265A CN 116702609 A CN116702609 A CN 116702609A
- Authority
- CN
- China
- Prior art keywords
- wire
- temperature
- model
- arima
- sag
- 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
Links
- 229920000049 Carbon (fiber) Polymers 0.000 title claims abstract description 48
- 239000004917 carbon fiber Substances 0.000 title claims abstract description 48
- 239000002131 composite material Substances 0.000 title claims abstract description 48
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000008859 change Effects 0.000 claims abstract description 10
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims abstract 8
- 230000005484 gravity Effects 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 claims description 4
- 230000001788 irregular Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000001932 seasonal effect Effects 0.000 claims description 4
- 238000013112 stability test Methods 0.000 claims description 3
- 238000001744 unit root test Methods 0.000 claims description 3
- 230000035882 stress Effects 0.000 description 41
- 230000005540 biological transmission Effects 0.000 description 11
- 238000012544 monitoring process Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 description 2
- 229910052782 aluminium Inorganic materials 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 241001123248 Arma Species 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012353 t test Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- 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
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
-
- 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
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- 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
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- 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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/16—Cables, cable trees or wire harnesses
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- 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)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Molecular Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Mathematical Optimization (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Geometry (AREA)
- Game Theory and Decision Science (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Hardware Design (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
Abstract
The invention discloses a carbon fiber composite core wire sag prediction method based on an ARIMA-LSTM combination algorithm, which comprises the steps of constructing an ARIMA model, predicting trend components in a temperature time sequence, adopting an LSTM model to predict residual components in the temperature time sequence, constructing another LSTM model to fit the relationship between an ARIMA model predicted value, a predicted value of a residual part sequence obtained by fitting the LSTM model and an acquired carbon fiber composite core wire temperature time sequence, establishing a wire temperature short-term predicted model, applying the wire temperature short-term predicted model to the monitored wire temperature time sequence, obtaining a future wire temperature predicted value in a short term, accurately predicting wire sag change by combining wire parameters, a wire stress state equation and a catenary equation, obtaining a predicted value of the wire sag in the short term, and using the predicted value in dynamic capacity regulation and control of a carbon fiber composite core overhead wire.
Description
Technical Field
The invention relates to the technical field of dynamic capacity expansion of power transmission lines, in particular to a carbon fiber composite core wire sag prediction method based on an ARIMA-LSTM combination algorithm.
Background
Along with the continuous rise of the electricity consumption load, under the condition that the existing transmission line is not changed, the improvement of the transmission capacity is a great problem which needs to be solved urgently by a power grid system, and the safety and stability of the power grid system and the healthy and rapid development of national economy construction are related.
The transmission capacity is improved on the basis of the existing transmission line, and the current method is mainly divided into a dynamic capacity-increasing method and a static capacity-increasing method. The main method of static capacity-increasing includes replacing and upgrading existing transmission line and installing new transmission line. The carbon fiber composite core compatibilized wire has low loss, high strength and good structural stability, and the existing steel-cored aluminum stranded wire is replaced, so that the redundant power transmission corridor is not increased, and the investment of funds is reduced, and therefore, the problem of hot spot of the compatibilized wire research is gradually solved. The dynamic capacity increasing method dynamically adjusts and predicts the maximum current-carrying capacity of the transmission line according to the temperature, sag and real-time change of the external environment condition of the line. The existing research mostly adopts transient and steady state heat balance equations to establish a current-carrying capacity physical model, and the current standard wire current-carrying capacity calculation in China mainly adopts a molar root formula. However, because the carbon fiber composite core capacity-increasing lead is applied less at present, the determination of some key performances and partial parameters of the carbon fiber composite core capacity-increasing lead is kept in a theoretical research level, the guarantee of actual operation is lacking, and meanwhile, the maximum allowable temperature of a steel-cored aluminum stranded wire is regulated to be 70 ℃ by the current standard, but the carbon fiber composite core capacity-increasing lead can operate for a long time at a high temperature of 160 ℃, so that the possible error of carrying capacity prediction of the carbon fiber composite core capacity-increasing lead is larger by adopting a traditional empirical formula. Because the carbon fiber composite core capacity-increasing lead has high allowable running temperature, the change of sag becomes an important index for dynamic capacity-increasing regulation and control of current-carrying capacity. Meanwhile, the change of sag is directly related to the temperature of the wire, so that accurate calculation and short-term prediction of the sag of the wire are of great significance for dynamic regulation and control of the capacity of the carbon fiber composite core capacity-increasing wire.
There are many devices and methods for calculating or monitoring sag of a power transmission line in the prior art to monitor or calculate sag in real time, but short-term prediction of sag still lacks relevant research. The change process of sag is an important index of dynamic capacity-increasing regulation of the capacity-increasing lead of the carbon fiber composite core, and accurate prediction of sag of the lead in a short period is an important research content for guaranteeing the reliability of dynamic capacity-increasing regulation of the capacity-increasing lead.
Disclosure of Invention
Aiming at the defects, the invention provides a carbon fiber composite core wire sag prediction method based on an ARIMA-LSTM combination algorithm, which can solve the problem that a sag prediction method is lacked in the prior art.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a carbon fiber composite core wire sag prediction method based on an ARIMA-LSTM combination algorithm comprises the following steps:
step S1, selecting a dynamic capacity-increasing regulation point of a carbon fiber composite core wire, acquiring a wire temperature time sequence, taking the acquired wire temperature time sequence as a training sample, and decomposing the training sample according to an addition model to obtain a trend component T and a residual part R of the sequence; determining parameters used by an ARIMA model, constructing the ARIMA model, and predicting a trend component T in a temperature time sequence to obtain an ARIMA model predicted value T 1 ；
S2, predicting the residual component R in the temperature time sequence in the step S1 by adopting an LSTM model to obtain an LSTM model predicted value R 1 ；
Step S3, constructing another LSTM model to fit the relationship between the ARIMA model predicted value in the step S1, the predicted value of the residual sequence obtained by LSTM model fitting in the step S2 and the acquired carbon fiber composite core wire temperature time sequence, and constructing a wire temperature short-term predicted model;
s4, applying a prediction model to the monitored wire temperature time sequence to obtain a predicted value of the future wire temperature in a short period; and combining the related parameters of the wire and a wire stress state equation, calculating to obtain the horizontal stress sigma of the wire at the predicted temperature, and substituting the horizontal stress sigma into a wire catenary equation to calculate to obtain the wire sag at the predicted temperature so as to obtain the predicted value of the wire sag in a short period.
Further, in step S4, the calculation of the horizontal stress σ of the wire at the predicted temperature and the sag of the wire at the measured temperature includes the following steps:
s41, obtaining wire parameters;
s42, aiming at the selected dynamic capacity-increasing regulation point, the temperature of the wire at the end is measured to be t 0 Horizontal stress sigma of wire at time 0 Taking the stress as a reference horizontal stress calculated later;
s43, obtaining the wire temperature t through a prediction model 1 Then, the elastic modulus, the gravity specific load, the temperature expansion coefficient and the temperature t of the lead are combined 0 Reference horizontal stress sigma 0 Substituting the stress state equation of the wire to calculate the horizontal stress sigma of the wire 1 ；
S44, horizontal stress sigma of the lead 1 Substituting the wire catenary equation to calculate and obtain the wire sag at the predicted temperature.
Further, the wire parameters in step S41 include the span, the height difference, the thermal expansion coefficient, the elastic modulus, and the gravity specific load of the wire.
Further, the wire stress state equation is:
wherein: alpha is the temperature expansion coefficient of the overhead line; e is the elastic modulus of the overhead line; sigma (sigma) 0 Initial horizontal stress when the installation of the overhead line is completed; t is the temperature of the wire in the current state; t is t 0 Is the initial temperature; sigma is the horizontal stress after the current overhead line state is changed; gamma is the ratio of the gravity of the wire per unit length and the sectional area after the state change; l is the gear distance.
Further, the catenary equation is:
in the formula, h is the height difference.
Further, the calculation formula of the maximum sag of the wire at the predicted temperature is:
further, the ARIMA model parameter determination in step S1 includes the following steps:
s11, performing stability test on the strain time sequence by adopting an ADF unit root test method, and processing the strain time sequence into a stable time sequence if the strain time sequence is judged to be non-stable;
s12, judging the model type through observing the tail and tail cutting conditions of the autocorrelation diagrams and the partial autocorrelation diagrams of the sequences, and preliminarily determining the value of the ARIMA model parameters.
Further, the acquisition of the temperature time sequence of the carbon fiber composite core wire is obtained by installing a temperature sensor at the end part of the wire.
Further, in step S1, performing time series decomposition on the training samples according to the addition model includes performing moving average on the original time series under a fixed period by using statsmodel toolkit, eliminating seasonal variation and irregular variation, obtaining a trend component T of the series, and after removing the trend component T from the original time series, uniformly calling the obtained time series as a remainder R.
Compared with the prior art, the invention has the beneficial effects that: according to the carbon fiber composite core wire sag prediction method based on the ARIMA-LSTM combination algorithm, an ARIMA model is built, trend components in a temperature time sequence are predicted, an LSTM model is adopted to predict residual components in the temperature time sequence, another LSTM model is built to fit the relationship between an ARIMA model predicted value, a predicted value of a residual part sequence obtained by fitting the LSTM model and an acquired carbon fiber composite core wire temperature time sequence, a wire temperature short-term predicted model is built, the wire temperature short-term predicted model is applied to the monitored wire temperature time sequence, future wire temperature predicted values in a short term are obtained, wire sag predicted values in a short term are obtained by accurately predicting wire sag changes in combination with wire parameters, a wire stress state equation and a catenary equation, and the predicted values can be used for dynamic capacity-increasing regulation of a carbon fiber composite core overhead wire.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are required to be used in the description of the embodiments will be briefly described below.
FIG. 1 is a flow chart of a carbon fiber composite core wire sag prediction method based on ARIMA-LSTM combination algorithm of the invention;
fig. 2 is a flow chart of sag calculation corresponding to the predicted temperature in step S4 of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. 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.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.
In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art. In addition, the technical features of the different embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
Taking a certain carbon fiber composite core capacity-increasing lead test line as an example, a sag short-term prediction method is described in detail. FIG. 1 is a schematic overall flow chart of a carbon fiber composite core wire sag prediction method based on an ARIMA-LSTM combination algorithm, as shown in FIG. 1, wherein Y (k) and Y (k-1) … Y (k-m) are wire temperature original time sequences actually measured by a temperature sensor, and m+1 data points are taken as a whole; t (T) 1 (k)、T 1 (k-1)…T 1 (k-m) is the result of the fitting of the ARIMA prediction model to the linear part (trend component T) of the original time series; r is R 1 (k)、R 1 (k-1)…R 1 (k-m) is a fitting result after predicting a nonlinear portion (remainder R) in the original time series in step S1; y ' (k) and Y ' (k+1) … Y ' (k+h) are prediction results obtained by the above combined prediction model, and total h+1 data points.
The invention discloses a carbon fiber composite core wire sag prediction method based on an ARIMA-LSTM combination algorithm, which specifically comprises the following steps:
step S1, selecting a dynamic capacity-increasing regulation point of a carbon fiber composite core wire, acquiring a wire temperature time sequence, taking the acquired wire temperature time sequence as a training sample, and decomposing the training sample according to an addition model to obtain a trend component T and a residual part R of the sequence; determining parameters used by an ARIMA model, constructing the ARIMA model, and predicting a trend component T in a temperature time sequence to obtain an ARIMA model predicted value T 1 。
Aiming at the carbon fiber composite core capacity-increasing lead test line, a large side of a certain tower is selected as a dynamic capacity-increasing regulation point, the span of the gear lead is 290m, and the height difference is about 20m. And obtaining parameters such as the span, the height difference, the thermal expansion coefficient, the elastic modulus, the gravity specific load and the like of the gear lead.
And a wire temperature sensor is arranged at the position of the wire at a position which is about 1m away from the wire hanging point, and the temperature of the wire is monitored continuously in real time. The data acquisition interval time is 5min, and the acquired data is transmitted to a background server in real time through a wireless public network (LTE/GPRS).
And acquiring the obtained wire temperature time sequence as historical data, taking the historical wire temperature monitoring data as a training sample, and carrying out time sequence decomposition on the training sample according to an addition model to obtain a trend component T (linear part in the time sequence) and a residual part R (nonlinear part in the time sequence) of the sequence. Determining parameters used by an ARIMA model, constructing the ARIMA model to predict a trend component T in a temperature time sequence, and obtaining an ARIMA model predicted value T 1 。
Specifically, the time series is first decomposed for training samples Y in the wire temperature actual measurement data. In general, the time series may decompose out a trend component T, a seasonal component S, a cyclic variation component C, and an irregular fluctuation component I. The composition of the four parts can be divided into three types of addition model, multiplication model and mixed model. The addition model can be expressed as: y=t+s+c+i; the multiplication model is expressed as: y=t×s×c×i; the hybrid model represents both plus and multiply signs in the formula.
The invention uses a common addition model and uses the statsmodel toolkit to decompose the original time series Y into a trend component T and a remainder R. The decomposition principle is as follows: and carrying out moving average on the original number sequence under a fixed period, eliminating seasonal variation and irregular variation, and obtaining a trend component T of the sequence. After the trend component T is removed from the original time series Y, the obtained time series are collectively referred to as a remainder R.
For the trend component T in the time series, which is considered as a linear part, an ARIMA (autoregressive integral moving average) prediction model is established to predict it. The ARIMA model basically adopts a plurality of differences to make a non-stationary sequence become a stationary sequence, and the differencesThe times are parameters d, the stable sequence is modeled by using an ARMA model taking p and q as parameters, and then the original sequence is obtained through inverse transformation. Firstly, carrying out stability test on data, and adopting an ADF (Augmented Dickey-Fuller) unit root test method to test a sequence to obtain a p value smaller than 0.05 and a t test value smaller than 1% critical value, wherein the time sequence is considered to be stable without carrying out differential and other transformations. And observing an Autocorrelation Chart (ACF) and a Partial Autocorrelation Chart (PACF) of the original sequence, and determining the best fit model correlation parameters of the original sequence. Obtaining the original time sequence linear part predictive value T by adopting ARIMA predictive model 1 。
S2, predicting the residual component R in the temperature time sequence in the step S1 by adopting an LSTM model to obtain an LSTM model predicted value R 1 。
Aiming at the residual part R obtained by the original time sequence decomposition, an LSTM model is established for fitting to obtain the predicted value R of the residual part 1 . Specifically, the ARIMA model is a linear model, and nonlinear components of the sequence are difficult to process, so that a long-short-term memory neural network (LSTM) model is required to be utilized to fit a nonlinear part R of a training sample to obtain a predicted value R of a residual sequence 1 Thereby improving the accuracy of the predictive model.
And S3, constructing another LSTM model to fit the relationship between the ARIMA model predicted value in the step S1, the predicted value of the residual sequence obtained by LSTM model fitting in the step S2 and the acquired carbon fiber composite core wire temperature time sequence, and establishing a wire temperature short-term predicted model.
Specifically, the following methods are adopted for time series prediction by using a common mixed model: considering that the linear model cannot process nonlinear components of the sequence, the fitted residual error contains nonlinear correlation of the time sequence, so that fitting and prediction are respectively carried out on a linear part and a nonlinear part in the original time sequence, and the predicted values of the two models are overlapped to obtain a final result. However, one drawback to this approach is that the model essentially assumes that the linear and nonlinear components of the sequence are simply linearly additive. In practice this is a simplification of the complex relationship of the two components, with a certain risk. The invention provides a new construction mode, which does not assume the relation between two components of the sequence, but utilizes a model with self-learning capability to fit the relation, thus reflecting the relation between the two components more essentially.
S4, applying a prediction model to the monitored wire temperature time sequence to obtain a predicted value of the future wire temperature in a short period; and combining the related parameters of the wire and a wire stress state equation, calculating to obtain the horizontal stress sigma of the wire at the predicted temperature, and substituting the horizontal stress sigma into a wire catenary equation to calculate to obtain the wire sag at the predicted temperature so as to obtain the predicted value of the wire sag in a short period.
In step S4, the calculation of the horizontal stress σ of the wire at the predicted temperature and the sag of the wire at the measured temperature includes the following steps:
s41, obtaining wire parameters, wherein the wire parameters comprise the span, the height difference, the thermal expansion coefficient, the elastic modulus and the gravity specific load of the wire;
s42, aiming at the selected dynamic capacity-increasing regulation point, the temperature of the wire at the end is measured to be t 0 Horizontal stress sigma of wire at time 0 Taking the stress as a reference horizontal stress calculated later;
s43, obtaining the wire temperature t through a prediction model 1 Then, the elastic modulus, the gravity specific load, the temperature expansion coefficient and the temperature t of the lead are combined 0 Reference horizontal stress sigma 0 Substituting the stress state equation of the wire to calculate the horizontal stress sigma of the wire 1 ；
S44, horizontal stress sigma of the lead 1 Substituting the wire catenary equation to calculate and obtain the wire sag at the predicted temperature.
In particular, the stiffness of the wires is often computationally ignored in the transmission line construction engineering design and is considered as a suspended flexible cable. The method for calculating the sag, the line length and the tension of the overhead line according to the principle that the unit length of the overhead line is uniformly distributed along the self line by gravity is called a catenary method for short. The catenary equation is:
wherein l is the horizontal distance (namely the span) of the wire hanging points on the two sides; h is the vertical distance (namely, height difference) between the wire hanging points on the two sides; gamma is the ratio of the gravity of the wire per unit length to the cross-sectional area (i.e. the specific load); sigma is the wire horizontal stress.
The maximum sag of the wire at the predicted temperature can be obtained by the above method:
as can be seen from the above equation, the maximum sag of the wire is only related to the horizontal stress σ and the specific load γ of the wire when the wire gauge l and the height difference h are determined.
For the horizontal stress sigma of the wire, the stress is inversely related to sag change, and when sag is larger, the stress is smaller, and conversely, the sag is smaller, and the stress is larger. In the process of changing the current-carrying capacity of the wire, the temperature of the wire is changed, and then the horizontal stress and sag of the wire are influenced. The wire stress state equation is:
wherein: alpha is the temperature expansion coefficient of the overhead line and DEG C -1 The method comprises the steps of carrying out a first treatment on the surface of the E is the elastic modulus of the overhead line, N/mm 2 ；σ 0 Initial horizontal stress when the installation of the overhead line is completed; t is the temperature of the wire in the current state; t is t 0 The sag design temperature is chosen to be the initial temperature, and is generally 40 ℃; sigma is the horizontal stress after the current overhead line state is changed; gamma is the ratio (specific load) of the gravity of the wire per unit length and the sectional area after the state change; l is the gear distance.
Considering that the aging and breaking force of the circuit are reduced along with the increase of the service life of the circuit, the calculation of the horizontal stress is also error, and the creep deformation of the lead can occur. Therefore, the reference horizontal stress σ in the wire stress state equation (3) used in the present invention 0 And reference temperature t 0 Or may be based on actual measurement values.
The established temperature time sequence prediction model is applied to actual monitoring data, prediction rolling is carried out, and the number of data points is predicted, namely h in the figure 1 is taken to be 2, namely wire temperatures of 5min and 10min in the future are predicted each time.
And after predicting the wire temperature, substituting the wire temperature into the formula (3) to obtain the horizontal stress sigma of the wire at the current temperature. Substituting the maximum sag into the equation (2) to obtain the maximum sag of the lead at the current temperature. The measured sag results within 20 minutes were compared with the real-time rolling prediction results of the present invention, as shown in table 1.
TABLE 1 comparison of sag predictions with actual values
| Time | Actual monitoring value/m | 5min prediction/m | 10min prediction/m |
| 5min | 7.65 | 7.69 | 7.87 |
| 10min | 7.96 | 8.01 | 8.07 |
| 15min | 8.12 | 8.15 | 8.21 |
| 20min | 8.15 | 8.13 | 8.11 |
According to the comparison of the actual sag monitoring value and the predicted values of 5min and 10min at the same time, the predicted value of the sag of the carbon fiber composite core capacity-increasing wire test line predicted by the prediction method disclosed by the invention has small difference from the actual monitoring value, and the error is not more than 5%, so that the reliability and the practicability of the established carbon fiber composite core capacity-increasing wire sag prediction model are verified, and the carbon fiber composite core capacity-increasing wire sag prediction model can be further applied to the dynamic capacity-increasing regulation and control of the carbon fiber composite core overhead wire.
The arc sag prediction method for the carbon fiber composite core wire based on the ARIMA-LSTM combination algorithm provided by the invention can be used for realizing short-term prediction (within a few ten minutes) of the maximum arc sag of the wire at the regulating point of the carbon fiber composite core wire. The ARIMA model is a linear model that has difficulty in processing the nonlinear components of the sequence, while the LSTM model has advantages for processing the nonlinear components. Meanwhile, the predicted two-part time sequence is not a simple addition relation, so that another LSTM model is applied to fit the relation between the two-part predicted value and the actual value. The temperature of the wire is accurately predicted in real time through the ARIMA-LSTM combination algorithm, and the sag change of the wire is accurately predicted by combining the wire parameters, the wire stress state equation and the catenary equation, and the method can be used for dynamic capacity-increasing regulation and control of the carbon fiber composite core overhead wire.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present invention should be included in the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (9)
1. The carbon fiber composite core wire sag prediction method based on ARIMA-LSTM combination algorithm is characterized by comprising the following steps of:
step S1, selecting a dynamic capacity-increasing regulation point of a carbon fiber composite core wire, acquiring a wire temperature time sequence, taking the acquired wire temperature time sequence as a training sample, and decomposing the training sample according to an addition model to obtain a trend component T and a residual part R of the sequence; determining parameters used by an ARIMA model, constructing the ARIMA model, and predicting a trend component T in a temperature time sequence to obtain an ARIMA model predicted value T 1 ；
S2, predicting the residual component R in the temperature time sequence in the step S1 by adopting an LSTM model to obtain an LSTM model predicted value R 1 ；
Step S3, constructing another LSTM model to fit the relationship between the ARIMA model predicted value in the step S1, the predicted value of the residual sequence obtained by LSTM model fitting in the step S2 and the acquired carbon fiber composite core wire temperature time sequence, and constructing a wire temperature short-term predicted model;
s4, applying a prediction model to the monitored wire temperature time sequence to obtain a predicted value of the future wire temperature in a short period; and combining the related parameters of the wire and a wire stress state equation, calculating to obtain the horizontal stress sigma of the wire at the predicted temperature, and substituting the horizontal stress sigma into a wire catenary equation to calculate to obtain the wire sag at the predicted temperature so as to obtain the predicted value of the wire sag in a short period.
2. The method for predicting sag of a carbon fiber composite core wire based on an ARIMA-LSTM combining algorithm as recited in claim 1, wherein,
in step S4, the calculation of the horizontal stress σ of the wire at the predicted temperature and the sag of the wire at the measured temperature includes the following steps:
s41, obtaining wire parameters;
s42, aiming at the selected dynamic capacity-increasing regulation point, the temperature of the wire at the end is measured to be t 0 Time guideHorizontal line stress sigma 0 Taking the stress as a reference horizontal stress calculated later;
s43, obtaining the wire temperature t through a prediction model 1 Then, the elastic modulus, the gravity specific load, the temperature expansion coefficient and the temperature t of the lead are combined 0 Reference horizontal stress sigma 0 Substituting the stress state equation of the wire to calculate the horizontal stress sigma of the wire 1 ；
S44, horizontal stress sigma of the lead 1 Substituting the wire catenary equation to calculate and obtain the wire sag at the predicted temperature.
3. The method for predicting sag of a carbon fiber composite core wire based on an ARIMA-LSTM combining algorithm as recited in claim 2, wherein,
the wire parameters in step S41 include the span, height difference, thermal expansion coefficient, elastic modulus and gravity specific load of the wire.
4. A method for predicting sag of a carbon fiber composite core wire based on ARIMA-LSTM combining algorithm as set forth in any one of claims 1 to 3,
the wire stress state equation is:
wherein: alpha is the temperature expansion coefficient of the overhead line; e is the elastic modulus of the overhead line; sigma (sigma) 0 At a temperature t 0 Horizontal stress of the wire; t is the temperature of the wire in the current state; t is t 0 Is the initial temperature; sigma is the horizontal stress after the current overhead line state is changed; gamma is the ratio of the gravity of the wire per unit length and the sectional area after the state change; l is the gear distance.
5. The method for predicting sag of a carbon fiber composite core wire based on an ARIMA-LSTM combining algorithm as recited in claim 4, wherein,
the catenary equation is:
in the formula, h is the height difference.
6. The method for predicting sag of a carbon fiber composite core wire based on an ARIMA-LSTM combining algorithm as recited in claim 5, wherein,
the calculation formula of the maximum sag of the wire at the predicted temperature is as follows:
7. the method for predicting sag of a carbon fiber composite core wire based on an ARIMA-LSTM combining algorithm as recited in claim 1, wherein,
the determining of the ARIMA model parameters in the step S1 comprises the following steps:
s11, performing stability test on the strain time sequence by adopting an ADF unit root test method, and processing the strain time sequence into a stable time sequence if the strain time sequence is judged to be non-stable;
s12, judging the model type through observing the tail and tail cutting conditions of the autocorrelation diagrams and the partial autocorrelation diagrams of the sequences, and preliminarily determining the value of the ARIMA model parameters.
8. The method for predicting sag of a carbon fiber composite core wire based on an ARIMA-LSTM combining algorithm as recited in claim 1, wherein,
the acquisition of the temperature time sequence of the carbon fiber composite core wire is realized by installing a temperature sensor at the end part of the wire.
9. The method for predicting sag of a carbon fiber composite core wire based on an ARIMA-LSTM combining algorithm as recited in claim 1, wherein,
in step S1, performing time series decomposition on the training samples according to the addition model includes performing moving average on the original time series under a fixed period by using statsmode toolkit, eliminating seasonal variation and irregular variation, obtaining a trend component T of the series, and uniformly calling the obtained time series as a remainder R after removing the trend component T from the original time series.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310672265.5A CN116702609A (en) | 2023-06-08 | 2023-06-08 | ARIMA-LSTM combination algorithm-based carbon fiber composite core wire sag prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310672265.5A CN116702609A (en) | 2023-06-08 | 2023-06-08 | ARIMA-LSTM combination algorithm-based carbon fiber composite core wire sag prediction method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116702609A true CN116702609A (en) | 2023-09-05 |
Family
ID=87827134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310672265.5A Pending CN116702609A (en) | 2023-06-08 | 2023-06-08 | ARIMA-LSTM combination algorithm-based carbon fiber composite core wire sag prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116702609A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117829352A (en) * | 2023-12-19 | 2024-04-05 | 浙江大学 | Industrial industry energy consumption prediction method and system based on clustering algorithm and machine learning |
-
2023
- 2023-06-08 CN CN202310672265.5A patent/CN116702609A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117829352A (en) * | 2023-12-19 | 2024-04-05 | 浙江大学 | Industrial industry energy consumption prediction method and system based on clustering algorithm and machine learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11429092B2 (en) | Asset management method for power equipment | |
| Zhang et al. | Overhead line preventive maintenance strategy based on condition monitoring and system reliability assessment | |
| Teh et al. | Risk-based management of transmission lines enhanced with the dynamic thermal rating system | |
| US20200134516A1 (en) | Method for asset management of electric power equipment | |
| CN116702609A (en) | ARIMA-LSTM combination algorithm-based carbon fiber composite core wire sag prediction method | |
| CN110729813A (en) | Intelligent operation and maintenance and full life cycle management method for transformer substation and cloud management platform | |
| CN109324261A (en) | A kind of power distribution network cable run overheat method for prewarning risk and system | |
| CN114460445B (en) | Transformer aging unavailability evaluation method considering aging threshold and service life | |
| KR101943455B1 (en) | Asset management method for substation | |
| CN112070413B (en) | Reliability research method of traction power supply system based on small sample condition | |
| CN115561564B (en) | ARIMA sequence prediction method for dynamic current-carrying capacity of cable joint | |
| CN114323337A (en) | Cable conductor temperature prediction method and system considering historical data | |
| CN113537643A (en) | Internet-based power equipment life prediction algorithm | |
| CN110363326B (en) | Power grid 110kV line short-term load curve prediction method | |
| KR102067831B1 (en) | Asset management method and apparatus for substation based on reliability indices of power grid | |
| CN114595883A (en) | Oil-immersed transformer residual life personalized dynamic prediction method based on meta-learning | |
| CN112801533B (en) | Power system operation reliability assessment method considering uncertainty of decision dependence | |
| US20200097922A1 (en) | Asset management method for substation | |
| Mahoor et al. | Leveraging sensory data in estimating transformer lifetime | |
| CN110782126B (en) | DC charging pile operation reliability evaluation method for electric automobile | |
| KR102309979B1 (en) | Asset management method for electric power equipment | |
| CN114915261A (en) | Fault monitoring method and device for photovoltaic power station | |
| CN112001073A (en) | Reliability analysis research method of traction power supply system | |
| Yepes-Fernández et al. | Lifetime degradation study of batteries operating as community energy storage systems | |
| CN110826179A (en) | Intelligent substation relay protection real-time reliability prediction method |
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 |