CN107103184A - A kind of high-voltage cable joint temperature predicting method - Google Patents
A kind of high-voltage cable joint temperature predicting method Download PDFInfo
- Publication number
- CN107103184A CN107103184A CN201710193168.2A CN201710193168A CN107103184A CN 107103184 A CN107103184 A CN 107103184A CN 201710193168 A CN201710193168 A CN 201710193168A CN 107103184 A CN107103184 A CN 107103184A
- Authority
- CN
- China
- Prior art keywords
- temperature
- particle
- connector
- vector machine
- voltage cable
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of high-voltage cable joint temperature predicting method, this method comprises the following steps:1) using environment temperature, ambient humidity, sheath/core electric current than, as the training sample of input, setting up the connector temperature forecast model based on least square method supporting vector machine with cable connector historical temperature value;2) dynamic optimization is carried out to the regularization parameter C in least square method supporting vector machine and normalizing parameter σ using particle cluster algorithm, forms the connector temperature forecast model based on particle group optimizing least square method supporting vector machine;3) temperature prediction is carried out according to connector temperature forecast model and real-time environment temperature, ambient humidity, sheath/core electric current ratio and cable connector historical temperature value based on particle group optimizing least square method supporting vector machine, obtains the predicted value of high-voltage cable joint temperature.Compared with prior art, the present invention has the advantages that convergence is good, has higher precision of prediction and faster training speed.
Description
Technical field
The present invention relates to high voltage power cable field, more particularly, to a kind of high-voltage cable joint temperature predicting method.
Background technology
Power cable transmission system not only can be with beautifying city appearance, moreover it is possible to significantly save urban land resource, fully
Meet resources conservation, environment-friendly demand.Power cable is able to extensive use in urban distribution network, and demand rapidly increases
It is long.The weak link of cable system is cable connector, and field operation experiences show, the cable operation fault more than 90% occurs
Cable connector position.The defect existed inside cable connector will cause electric field to concentrate, local temperature rise, more than 137 DEG C, electric power
Cable insulation medium punctures occurring electric-thermal, has a strong impact on the safe and reliable operation of power network.Connector temperature can reflect electricity well
The operation conditions of cable joint, existing cable monitoring system can carry out real-time data acquisition to temperature of cable junction, but can not carry out
Forecast.Therefore, reply connector temperature is predicted, and the dielectric level of anticipation cable connector, timely failure judgement, are temperature in advance
Monitoring system provides early warning foundation.
The content of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide a kind of high-voltage cable joint
Temperature predicting method.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of high-voltage cable joint temperature predicting method, this method comprises the following steps:
1) using environment temperature, ambient humidity, sheath/core electric current than with cable connector historical temperature value as input instruction
Practice sample, set up the connector temperature forecast model based on least square method supporting vector machine;
2) the regularization parameter C in least square method supporting vector machine and normalizing parameter σ is carried out using particle cluster algorithm
Dynamic optimization, forms the connector temperature forecast model based on particle group optimizing least square method supporting vector machine;
3) according to connector temperature forecast model and real-time environment based on particle group optimizing least square method supporting vector machine
Temperature, ambient humidity, sheath/core electric current ratio and cable connector historical temperature value carry out temperature prediction, obtain high-tension cable and connect
The predicted value of head temperature.
Described step 1) in the connector temperature forecast model based on least square method supporting vector machine be:
K(x,xi)=exp (- | | x-xi||2/σ2)
Wherein, y is output vector, λiFor Lagrange multiplier, K (xi, it is x) kernel function, b is departure, xiFor i-th
The center of RBF, x is input vector, and l is training sample data point sum, and σ is normalizing parameter, | | x-xi| | be to
Measure x-xiNorm.
Described kernel function is Radial basis kernel function.
Described step 2) in, using particle cluster algorithm to the regularization parameter C and mark in least square method supporting vector machine
Standardization parameter σ carries out dynamic optimization and specifically includes following steps:
21) initialization aceleration pulse c1And c2, inertia weight w, population scale m, maximum evolutionary generation Nmax, given threshold ε,
And regularization parameter C and nuclear parameter σ are mapped as the population of input;
22) obtain each particle of current location adaptive value and make comparisons, in each particle select optimal location and
Most the superior is used as population optimal location in all particles;
23) speed and the position of each particle are updated, new population is produced;
24) adaptive value of the particle of each in new population new position, and respectively with its history optimal location and the history of population
Optimal location is made comparisons, if more excellent, is replaced, otherwise, is kept constant;
25) judge whether to meet termination condition, if meeting, export optimal C and σ, otherwise, return to step 22).
Described step 23) in, the speed v of j-th of particlejdWith position xjdRenewal rule be:
vjd=wvjd+c1r1(pjd-xjd)+c2r2(pgd-xjd)
xjd=xjd+vjd
Wherein, pjdFor the optimal location of j-th of particle, pgdFor the optimal location in all particles, r1、r2For between 0-1
Random number.
Described step 25) in, termination condition is that evolutionary generation reaches NmaxOr precision is less than given threshold ε.
Compared with prior art, the present invention has advantages below:
The present invention sets up connector temperature forecast model using LS-SVM first, with environment temperature, ambient humidity, sheath/line
Core electric current ratio and cable connector last temperature as training sample.In order to improve precision of prediction, using particle cluster algorithm
(particle swarm optimization, PSO) enters Mobile state to LS-SVM regularization parameter C and normalizing parameter σ and sought
It is excellent, build PSO-LSSVM Forecasting Methodologies.By taking the 110kV tags of Shanghai as an example, predict the outcome and show, this prediction side
Method convergence is good, have higher precision of prediction and faster training speed, and can be provided for cable temperature detection with early warning system can
The basis for estimation leaned on.
Brief description of the drawings
Fig. 1 is that temperature of cable junction predicts decorum structure.
Fig. 2 is the flow that PSO-LSSVM predicts temperature of cable junction.
Fig. 3 predicts the outcome for experiment 1, wherein, figure (3a) is the temperature prediction curve of experiment 1, and Fig. 3 (3b) is experiment 1
Relative error.
Fig. 4 predicts the outcome for experiment 2, wherein, figure (4a) is the temperature prediction curve of experiment 2, and Fig. 4 (4b) is experiment 2
Relative error.
Embodiment
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.
Embodiment:
1st, least square method supporting vector machine prediction algorithm
The basic thought of SVMs data regression is by the influence factor x closely related with premeasuringiIt is used as input
Amount, prediction desired value yiAs output quantity, pass through Nonlinear MappingHigh-dimensional feature space is mapped to from the input space, its is non-
Linear relationship is expressed as:
Data point set (xi,yi), i=1 ..., l, xi∈Rd,yi∈ R, d are the dimension of selected input variable, and l is datum
The sum at strong point, weight vector ω ∈ Rd, departure b ∈ R,It is that non-linear from the input space to high-dimensional feature space reflects
Penetrate.The secondary norm of error is chosen as loss function, by risk minimization principle, LS-SVM Optimized models are represented by:
In formula, eiFor error, e ∈ Rl×1For error vector, C is regularization parameter, controls the punishment degree to error.Draw
Enter Lagrange multipliers, λ ∈ Rl×1, formula (2) can be converted into:
By KKT conditions:
ω and e is eliminated, system of linear equations is obtained
Wherein, λ=[λ1,λ2,...λl]T, U=[1,1 ..., 1]T, Y=[y1,y2,...,yl]TFor the dimensional vector of l × 1,
Ω∈Rl×l, andK is the kernel function for meeting Mercer conditions.Conventional kernel function has
Linear kernel function, perceptron kernel function and Radial basis kernel function, wherein Radial basis kernel function better performances[9].Present invention selection footpath
To kernel function of the base kernel function as LS-SVM, form is as follows:
K(x,xi)=exp (- | | x-xi||2/σ2) (6)
Wherein:X is m dimensional input vectors, xiIt is the center of i-th of RBF, there is same dimension with x, σ is standard
Change parameter, determine that the function surrounds the width of central point, | | x-xi| | it is vector x-xiNorm, represent x and xiBetween away from
From.
So far, quadratic programming problem changes into Solving Linear, and the expression formula of LS-SVM forecast models is:
2nd, particle group optimizing LS-SVM parameters
When being predicted using LS-SVM, it is thus necessary to determine that the regularization parameter C of control punishment degree and the standardization of kernel function
Parameter σ.C obtains too small, then just small to sample punishment, training error is become big;C obtains excessive, and training error diminishes, then wide energy
Power is deteriorated.σ explications high-dimensional feature spaceStructure, control last solution complexity.σ is too small, easily localization
Training is crossed, σ is excessive, easily cause deficient training.
Parameter C and σ determination, are substantially the processes of a dynamic optimization.In order that LS-SVM has more preferable precision of prediction,
The present invention optimizes selection using particle swarm optimization algorithm to C and σ.Particle cluster algorithm asks the parameter selection of SVMs
Topic is considered as the global search problem in given space, the Rule of judgment terminated using the mean error of test sample collection as algorithm,
Realize that the Automatic Optimal of parameter is chosen.Comprise the following steps that:
Step 1 is to sample data normalized, initialization aceleration pulse c1And c2, inertia weight w, population scale m, most
Macroevolution algebraically Nmax, regularization parameter C and nuclear parameter σ are mapped as a group particle.
Step 2 calculates the adaptive value f (x of each particle of current locationi) and make comparisons, i-th of particle current point is set to
Optimal location pibest, most the superior is set to population optimal location g in all particlesbest。
Step 3 updates speed and the position of each particle according to formula (8) and (9), produces new population X (t).
vid=wvid+c1r1(pid-xid)+c2r2(pgd-xid) (8)
xid=xid+vid (9)
Step 4 calculates the adaptive value of X (t) each particle new position, and gone through respectively with its history optimal location and population
History optimal location is made comparisons, if more excellent, is replaced, otherwise, is kept constant.
Step 5 checks that termination condition (reaches NmaxOr precision is less than ε), terminate if meeting, export optimal C and σ, otherwise,
T=t+1 is made, 2 are gone to step.
3rd, temperature of cable junction is predicted
3.1 system architectures and data acquisition
Temperature of cable junction is relevant with environment temperature, humidity, load current, and load current has certain with circulating current
Relation, assign the ratio of circulating current and cable core as input sample.Model system structure is as shown in Figure 1.Environment temperature, ring
Border humidity, sheath/core electric current ratio and cable connector historical temperature are obtained by high-tension cable running status on-line monitoring system, often
Group data are by the average ambient temperature on the same day, highest environment temperature, minimum environment temperature, ambient humidity, sheath/core electric current
Than, joint observed temperature totally 29 data composition per hour, multi-group data is chosen as input sample, is exported by forecast model
Predicted temperature.
3.2nd, prediction steps and parameter are chosen
The average relative error MAPE and root mean square relative error MSE that present invention selection is commonly used in temperature prediction are as by mistake
The evaluation criterion of difference.
Relative error (RE-Relative Error)
Average relative error (MAPE-Mean Absolute Percentage Error)
L in formula,Respectively actual temperature and predicted temperature.N is the number of temperature data.
So far, it may be determined that the step of PSO-LSSVM predicts temperature of cable junction, as shown in Figure 2.
Program, the initialization of each parameters of PSO are write with Matlab:Population m takes 20, maximum iteration Nmax10 are taken, is used to
Property weight coefficient w spans be [0.4,0.9], in order to balance c in the effect of enchancement factor, the present invention1And c2It is 2.
3 days to 9 October in 2016 is taken as test object, respectively using No. 3 outdoor terminal joint A phases of Shanghai 110kV cables
The temperature (referred to as experiment 1) that the Monitoring Data of day predicts the 10th, takes the data prediction in 3 to 17 October in 2016 the 16th
Temperature (referred to as experiment 2).In order to verify the precision of prediction of PSO-LSSVM methods, experiment 1 and experiment 2 use LS-SVM simultaneously
Prediction is used as data comparison.Table 1 is that the Model Parameter Optimization result tested twice is peaceful with respect to application condition, in testing twice
LS-SVM parameter C and σ takes 30,2 respectively, and PSO-LSSVM parameter C and σ is optimized for 143.52 and 7.18 in experiment 1, real
The parameter C and σ for testing PSO-LSSVM in 1 are optimized for 7.0857 and 10.
The Model Parameter Optimization result of table 1 peace is with respect to application condition
The experiment of table 22 predicts the outcome
Fig. 3 (a) predicts the outcome for experiment 1, shown in relative error such as Fig. 3 (b), and Fig. 4 (a) predicts the outcome for experiment 2,
Fig. 4 (b) is its relative error, and table 2 gives the data that predict the outcome of experiment 2, it can be seen from Fig. 3, Fig. 4 and table 1, table 2
PSO-LSSVM prediction is superior to LS-SVM predictions closer to measured value, relative error and average relative error, and parameter optimization has
Effect.The average relative error of experiment 1 is 0.0553%, more than the 0.021% of experiment 2, and the size of this explanation data sample can shadow
Sound predicts the outcome, and Rational choice data sample is conducive to improving precision of prediction.The maximum temperature error of experiment 2 occurs at 12,
For 2.143 DEG C, absolute relative error is 0.0748%, meets prediction and requires.
The present invention carries out dynamic optimization using PSO to LS-SVM parameter, with joint historical temperature, environment temperature, humidity,
Core/circulating current is used for training sample, sets up PSO-LSSVM temperature of cable junction forecast models, improves estimated performance,
Temperature of cable junction can effectively be predicted.The size of data sample can influence to predict the outcome, and Rational choice data sample is conducive to carrying
High precision of prediction.Test result indicates that, this method has that convergence is good, have higher precision of prediction and faster training speed etc.
Advantage, can be that cable temperature detection and early warning system provide reliable basis for estimation, with very high engineering application value.
Claims (6)
1. a kind of high-voltage cable joint temperature predicting method, it is characterised in that this method comprises the following steps:
1) using environment temperature, ambient humidity, sheath/core electric current than with cable connector historical temperature value as input training sample
This, sets up the connector temperature forecast model based on least square method supporting vector machine;
2) Mobile state is entered to the regularization parameter C and normalizing parameter σ in least square method supporting vector machine using particle cluster algorithm
Optimizing, forms the connector temperature forecast model based on particle group optimizing least square method supporting vector machine;
3) according to based on particle group optimizing least square method supporting vector machine connector temperature forecast model and real-time environment temperature,
Ambient humidity, sheath/core electric current ratio and cable connector historical temperature value carry out temperature prediction, obtain high-voltage cable joint temperature
Predicted value.
2. a kind of high-voltage cable joint temperature predicting method according to claim 1, it is characterised in that described step 1)
In the connector temperature forecast model based on least square method supporting vector machine be:
<mrow>
<mi>y</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>l</mi>
</munderover>
<msub>
<mi>&lambda;</mi>
<mi>i</mi>
</msub>
<mi>K</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mi>b</mi>
</mrow>
K(x,xi)=exp (- | | x-xi||2/σ2)
Wherein, y is output vector, λiFor Lagrange multiplier, K (xi, it is x) kernel function, b is departure, xiFor i-th radially
The center of basic function, x is input vector, and l is training sample data point sum, and σ is normalizing parameter, | | x-xi| | for vector x-
xiNorm.
3. a kind of high-voltage cable joint temperature predicting method according to claim 1, it is characterised in that described kernel function
For Radial basis kernel function.
4. a kind of high-voltage cable joint temperature predicting method according to claim 1, it is characterised in that described step 2)
In, dynamic optimization is carried out to the regularization parameter C in least square method supporting vector machine and normalizing parameter σ using particle cluster algorithm
Specifically include following steps:
21) initialization aceleration pulse c1And c2, inertia weight w, population scale m, maximum evolutionary generation Nmax, given threshold ε, and just
Then change the population that parameter C and nuclear parameter σ is mapped as input;
22) obtain the adaptive value of each particle of current location and make comparisons, optimal location is selected in each particle and all
Most the superior is used as population optimal location in particle;
23) speed and the position of each particle are updated, new population is produced;
24) adaptive value of the particle of each in new population new position, and the history respectively with its history optimal location and population is optimal
Position is made comparisons, if more excellent, is replaced, otherwise, is kept constant;
25) judge whether to meet termination condition, if meeting, export optimal C and σ, otherwise, return to step 22).
5. a kind of high-voltage cable joint temperature predicting method according to claim 4, it is characterised in that described step
23) in, the speed v of j-th of particlejdWith position xjdRenewal rule be:
vjd=wvjd+c1r1(pjd-xjd)+c2r2(pgd-xjd)
xjd=xjd+vjd
Wherein, pjdFor the optimal location of j-th of particle, pgdFor the optimal location in all particles, r1、r2For between 0-1 with
Machine number.
6. a kind of high-voltage cable joint temperature predicting method according to claim 4, it is characterised in that described step
25) in, termination condition is that evolutionary generation reaches NmaxOr precision is less than given threshold ε.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710193168.2A CN107103184A (en) | 2017-03-28 | 2017-03-28 | A kind of high-voltage cable joint temperature predicting method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710193168.2A CN107103184A (en) | 2017-03-28 | 2017-03-28 | A kind of high-voltage cable joint temperature predicting method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107103184A true CN107103184A (en) | 2017-08-29 |
Family
ID=59674834
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710193168.2A Pending CN107103184A (en) | 2017-03-28 | 2017-03-28 | A kind of high-voltage cable joint temperature predicting method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107103184A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108844624A (en) * | 2018-06-01 | 2018-11-20 | 北京科技大学 | A kind of SLM process laser power monitor method based on temperature field |
CN109285331A (en) * | 2018-11-29 | 2019-01-29 | 国网上海市电力公司 | A kind of Temperature of Power Cables early warning system based on data analysis and temperature prediction |
CN109615121A (en) * | 2018-11-27 | 2019-04-12 | 西安理工大学 | Bullet train axle temperature predicting method based on data-driven support vector machines |
CN109870627A (en) * | 2017-12-05 | 2019-06-11 | 华北电力大学(保定) | Submarine cable fault alarm and diagnostic method based on distributed fiber optic temperature strain and vibration monitoring data |
CN110705813A (en) * | 2019-07-23 | 2020-01-17 | 电子科技大学 | Hybrid cable connection method considering reliability of wind power plant current collection system |
CN110907064A (en) * | 2019-11-20 | 2020-03-24 | 国网重庆市电力公司电力科学研究院 | GIS disconnecting switch contact temperature prediction method and device and readable storage medium |
CN111382895A (en) * | 2018-12-29 | 2020-07-07 | 中国电力科学研究院有限公司 | Seawater desalination load prediction method and system |
CN112013993A (en) * | 2020-08-27 | 2020-12-01 | 国网山西省电力公司大同供电公司 | Submarine cable detection method based on underwater robot |
CN115577643A (en) * | 2022-11-23 | 2023-01-06 | 广东电网有限责任公司中山供电局 | Temperature prediction method and device for cable terminal |
CN115796057A (en) * | 2023-02-06 | 2023-03-14 | 广东电网有限责任公司中山供电局 | Cable joint temperature prediction method and system based on BAS-LSTM |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101464488A (en) * | 2009-01-16 | 2009-06-24 | 上海慧东电气设备有限公司 | On-line monitoring system for high voltage cable |
CN101788593A (en) * | 2010-01-11 | 2010-07-28 | 山东康威通信技术股份有限公司 | State monitoring and analysis system of power cable sheath grounding current and method thereof |
CN103886374A (en) * | 2014-04-22 | 2014-06-25 | 武汉大学 | Cable joint wire temperature prediction method based on RBF neural network |
CN104361393A (en) * | 2014-09-06 | 2015-02-18 | 华北电力大学 | Method for using improved neural network model based on particle swarm optimization for data prediction |
-
2017
- 2017-03-28 CN CN201710193168.2A patent/CN107103184A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101464488A (en) * | 2009-01-16 | 2009-06-24 | 上海慧东电气设备有限公司 | On-line monitoring system for high voltage cable |
CN101788593A (en) * | 2010-01-11 | 2010-07-28 | 山东康威通信技术股份有限公司 | State monitoring and analysis system of power cable sheath grounding current and method thereof |
CN103886374A (en) * | 2014-04-22 | 2014-06-25 | 武汉大学 | Cable joint wire temperature prediction method based on RBF neural network |
CN104361393A (en) * | 2014-09-06 | 2015-02-18 | 华北电力大学 | Method for using improved neural network model based on particle swarm optimization for data prediction |
Non-Patent Citations (3)
Title |
---|
朱霄珣 等: "基于PSO参数优化的LS-SVM风速预测方法研究", 《中国电机工程学报》 * |
田成凤: "电力电缆接头的温度监测与预警研究", 《电线电缆》 * |
肖微 等: "基于广义回归神经网络的电缆接头温度预测", 《电气安全》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109870627B (en) * | 2017-12-05 | 2021-06-18 | 华北电力大学(保定) | Submarine cable fault alarming and diagnosing method based on distributed optical fiber temperature strain and vibration monitoring data |
CN109870627A (en) * | 2017-12-05 | 2019-06-11 | 华北电力大学(保定) | Submarine cable fault alarm and diagnostic method based on distributed fiber optic temperature strain and vibration monitoring data |
CN108844624B (en) * | 2018-06-01 | 2020-01-21 | 北京科技大学 | SLM process laser power monitoring method based on temperature field |
CN108844624A (en) * | 2018-06-01 | 2018-11-20 | 北京科技大学 | A kind of SLM process laser power monitor method based on temperature field |
CN109615121B (en) * | 2018-11-27 | 2024-01-23 | 西安理工大学 | High-speed train axle temperature prediction method based on data driving support vector machine |
CN109615121A (en) * | 2018-11-27 | 2019-04-12 | 西安理工大学 | Bullet train axle temperature predicting method based on data-driven support vector machines |
CN109285331A (en) * | 2018-11-29 | 2019-01-29 | 国网上海市电力公司 | A kind of Temperature of Power Cables early warning system based on data analysis and temperature prediction |
CN111382895A (en) * | 2018-12-29 | 2020-07-07 | 中国电力科学研究院有限公司 | Seawater desalination load prediction method and system |
CN110705813A (en) * | 2019-07-23 | 2020-01-17 | 电子科技大学 | Hybrid cable connection method considering reliability of wind power plant current collection system |
CN110705813B (en) * | 2019-07-23 | 2022-11-22 | 电子科技大学 | Hybrid cable connection method considering reliability of wind power plant current collection system |
CN110907064A (en) * | 2019-11-20 | 2020-03-24 | 国网重庆市电力公司电力科学研究院 | GIS disconnecting switch contact temperature prediction method and device and readable storage medium |
CN112013993A (en) * | 2020-08-27 | 2020-12-01 | 国网山西省电力公司大同供电公司 | Submarine cable detection method based on underwater robot |
CN115577643A (en) * | 2022-11-23 | 2023-01-06 | 广东电网有限责任公司中山供电局 | Temperature prediction method and device for cable terminal |
CN115796057A (en) * | 2023-02-06 | 2023-03-14 | 广东电网有限责任公司中山供电局 | Cable joint temperature prediction method and system based on BAS-LSTM |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107103184A (en) | A kind of high-voltage cable joint temperature predicting method | |
CN110311376B (en) | Dynamic safety assessment comprehensive model and space-time visualization method for power system | |
CN103972884B (en) | A kind of power system state estimation method | |
CN113964885B (en) | Situation awareness-based active power grid reactive power prediction and control method | |
CN111525587B (en) | Reactive load situation-based power grid reactive voltage control method and system | |
CN112070413B (en) | Reliability research method of traction power supply system based on small sample condition | |
CN110969306A (en) | Power distribution low-voltage distribution area load prediction method and device based on deep learning | |
LU500551B1 (en) | Virtual load dominant parameter identification method based on incremental learning | |
CN114742283A (en) | Medium-voltage line loop closing current estimation and risk assessment method based on deep learning network | |
CN116756530B (en) | Power grid line loss evaluation method and system for new energy access power distribution network | |
CN110045237A (en) | Transformer state parametric data prediction technique and system based on drosophila algorithm optimization | |
JP2023538611A (en) | Method and computer system for generating decision logic for a controller | |
CN112649695A (en) | Power quality assessment scheme for power distribution network based on node full coverage | |
CN103155331A (en) | Method and device for producing a state signal | |
CN114707422A (en) | Intelligent power check method based on load prediction | |
CN107742886B (en) | Prediction method for load peak simultaneous coefficient of thermoelectric combined system | |
CN105552906B (en) | A kind of area power grid load nargin analysis method based on prim al- dual interior point m ethod | |
Moradzadeh et al. | Image processing-based data integrity attack detection in dynamic line rating forecasting applications | |
He | Temperature Prediction of Power Cable Joint Based on LS-SVM Optimized by PSO | |
CN117277446B (en) | Multi-target power distribution network planning method and system | |
Bai et al. | Abnormal Detection Scheme of Substation Equipment based on Intelligent Fusion Terminal | |
CN110633794A (en) | Elman neural network-based high-voltage cable conductor temperature dynamic calculation method | |
Song et al. | Online Life Forecasting Method and State Assessment Architecture of Power Cable Based on Recurrent Neural Network and Internet of Things | |
Chen et al. | Secure Probabilistic Interval Prediction of Dynamic Thermal Rating Against Weather Imbalance Constraints | |
Wang et al. | The residual life prediction of power grid transformers based on GA-ELM computational model and digital twin data |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170829 |
|
RJ01 | Rejection of invention patent application after publication |