CN114323337A - Cable conductor temperature prediction method and system considering historical data - Google Patents
Cable conductor temperature prediction method and system considering historical data Download PDFInfo
- Publication number
- CN114323337A CN114323337A CN202111449826.2A CN202111449826A CN114323337A CN 114323337 A CN114323337 A CN 114323337A CN 202111449826 A CN202111449826 A CN 202111449826A CN 114323337 A CN114323337 A CN 114323337A
- Authority
- CN
- China
- Prior art keywords
- conductor
- current
- cable
- historical
- temperature
- 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.)
- Granted
Links
- 239000004020 conductor Substances 0.000 title claims abstract description 257
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000012544 monitoring process Methods 0.000 claims abstract description 93
- 238000012549 training Methods 0.000 claims abstract description 44
- 238000000547 structure data Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims description 47
- 238000009413 insulation Methods 0.000 claims description 30
- 238000005457 optimization Methods 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000004364 calculation method Methods 0.000 description 7
- 230000008859 change Effects 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000004088 simulation Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 229920003020 cross-linked polyethylene Polymers 0.000 description 1
- 239000004703 cross-linked polyethylene Substances 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000002922 simulated annealing Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Landscapes
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The invention provides a cable conductor temperature prediction method and a system considering historical data, which are used for obtaining a cable conductor temperature prediction model by pre-training and comprise the following steps: a1, collecting current and historical monitoring data and corresponding structural data of the high-voltage cable; step A2, taking the cable skin temperature, the cable conductor current in the historical monitoring data, the skin temperature, the conductor current and the structural data in the current monitoring data as input, and taking the conductor temperature as output to obtain a cable conductor temperature prediction model; s1, collecting the current and historical conductor monitoring data and cable structure data of the cable to be predicted; and step S2, inputting the current and historical conductor monitoring data and the cable structure data into a cable conductor temperature prediction model to obtain a conductor temperature prediction value. The method and the system have the beneficial effects that the predicted value of the conductor temperature is accurately calculated by taking the cable skin temperature, the historical time value and the current time value of the cable conductor current into account.
Description
Technical Field
The invention relates to the technical field of cable conductor temperature prediction, in particular to a cable conductor temperature prediction method and system considering historical data.
Background
The high-voltage power cables are the main transmission mode of electric energy among all regions, and because the advantages of small occupied space and small maintenance workload, the number of the high-voltage power cables is increasing day by day, the current carrying capacity of the cables is limited by conductor temperature, the service life of the cables can be shortened by the high conductor temperature, cable resource waste can be caused by the low conductor temperature, and therefore the conductor temperature of the cables can be accurately obtained, and the method has important significance for evaluating the running state of the cables and improving the transmission capacity.
The existing conductor temperature prediction method can calculate the conductor temperature of a cable under given conditions, but the accuracy of the result depends heavily on the accurate setting of initial conditions, the method is not suitable for the real-time prediction of the conductor temperature of the cable on site, the conductor temperature prediction based on the current state quantity and a steady-state thermal circuit model brings severe leading errors, the prediction is carried out based on a transient model, and the accuracy of the prediction result also depends heavily on the accurate setting of the initial conditions.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a cable conductor temperature prediction method considering historical data, which is trained in advance to obtain a cable conductor temperature prediction model and specifically comprises the following steps:
step A1, acquiring current monitoring data, at least one group of historical monitoring data and corresponding structural data of at least one high-voltage cable, wherein the current monitoring data comprises a skin temperature, a conductor current and a conductor temperature at the current data acquisition moment, and each group of historical monitoring data comprises a cable skin temperature and a cable conductor current at a plurality of historical data acquisition moments;
step A2, taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current and the structural data of each historical data acquisition moment as input, taking the conductor temperature as output, and training to obtain a cable conductor temperature prediction model;
the method for predicting the temperature of the cable conductor specifically comprises the following steps:
step S1, collecting current conductor monitoring data of a cable to be predicted at the current collecting moment, historical conductor monitoring data of at least one historical collecting moment and corresponding cable structure data;
and step S2, inputting the current conductor monitoring data, the historical conductor monitoring data and the cable structure data into the cable conductor temperature prediction model to obtain a conductor temperature prediction value.
Preferably, the current conductor monitoring data includes a current cable skin temperature and a current cable conductor current at a current collecting time, and the historical conductor monitoring data includes a historical cable skin temperature and a historical cable conductor current corresponding to at least one historical collecting time before the current collecting time.
Preferably, the structural data includes an insulation layer equivalent thermal resistance value, an outer sheath equivalent thermal resistance value, an insulation layer loss value and an outer sheath loss value of the high-voltage cable.
Preferably, the step a2 includes:
step A21, processing according to the skin temperature of the cable at each historical data acquisition time, the skin temperature at the current data acquisition time and a preset weight factor to obtain a skin temperature equivalent value, and processing according to the conductor current of the cable at each historical data acquisition time, the conductor current at the current data acquisition time and the weight factor to obtain a conductor current equivalent value;
step A22, processing according to the skin temperature equivalent value, the conductor current equivalent value, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value, the outer sheath loss value and the weight factor to obtain a temperature prediction expression;
step A23, based on the temperature prediction expression, taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value and the outer sheath loss value of each historical data acquisition moment as input, taking the conductor temperature as output, and training to obtain the cable conductor temperature prediction model.
Preferably, the weighting factor is optimized according to a heuristic optimization algorithm to obtain an optimized weighting factor, and then in step a22, the temperature prediction expression is obtained by processing according to the skin temperature equivalent value, the conductor current equivalent value, the insulating layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulating layer loss value, the outer sheath loss value, and the optimized weighting factor.
Preferably, a set of historical monitoring data is collected in the step a1, and then in the step S1, the historical conductor monitoring data includes the historical cable skin temperature and the historical cable conductor current corresponding to one historical collection time before the current collection time.
Preferably, if a plurality of sets of historical monitoring data are collected in the step a1, the historical conductor monitoring data includes the historical cable skin temperature and the historical cable conductor current corresponding to a plurality of historical collection times before the current collection time in the step S1.
Preferably, a cable conductor temperature prediction system taking historical data into account is applied to the cable conductor temperature prediction method, and includes:
a model training module, comprising:
the data acquisition unit is used for acquiring current monitoring data, at least one group of historical monitoring data and corresponding structural data of at least one high-voltage cable, wherein the current monitoring data comprises a skin temperature, a conductor current and a conductor temperature at the current data acquisition moment, and each group of historical monitoring data comprises a cable skin temperature and a cable conductor current at a plurality of historical data acquisition moments;
the model training unit is connected with the data acquisition unit and used for training the cable skin temperature, the cable conductor current, the skin temperature, the conductor current and the structural data at the current data acquisition time as input, and the conductor temperature as output to obtain the cable conductor temperature prediction model;
the data acquisition module is used for acquiring current conductor monitoring data of a cable to be predicted at the current acquisition moment, historical conductor monitoring data of at least one historical acquisition moment and corresponding cable structure data;
and the temperature prediction module is respectively connected with the data acquisition module and the model training module and is used for inputting the current conductor monitoring data, the historical conductor monitoring data and the cable structure data into the cable conductor temperature prediction model to obtain a conductor temperature prediction value.
Preferably, the structural data includes an insulation layer equivalent thermal resistance value, an outer sheath equivalent thermal resistance value, an insulation layer loss value, and an outer sheath loss value of the high-voltage cable, and the model training unit includes:
the first processing subunit is used for processing the cable skin temperature at each historical data acquisition moment, the skin temperature at the current data acquisition moment and a preset weight factor to obtain a skin temperature equivalent value, and processing the cable conductor current at each historical data acquisition moment, the conductor current at the current data acquisition moment and the weight factor to obtain a conductor current equivalent value;
the second processing subunit is connected with the first processing subunit and used for processing according to the skin temperature equivalent value, the conductor current equivalent value, the insulating layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulating layer loss value, the outer sheath loss value and the weight factor to obtain a temperature prediction expression;
and the model training subunit is connected with the second processing subunit and used for training to obtain the cable conductor temperature prediction model by taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value and the outer sheath loss value of each historical data acquisition moment as input and taking the conductor temperature as output based on the temperature prediction expression.
Preferably, the model training unit further includes a factor optimization unit, respectively connected to the first processing unit and the second processing unit, and configured to optimize the weight factor according to a heuristic optimization algorithm to obtain an optimized weight factor and control the first processing subunit and the second processing subunit to update the weight factor to the optimized weight factor.
The technical scheme has the following advantages or beneficial effects:
(1) the method and the system do not need to acquire environmental data of the environment except the cable, have less parameters needed by the model and are suitable for any complex environment;
(2) the method and the system consider that the response time of the cable thermal process is longer, and account for the cable skin temperature, the historical time value and the current time value of the cable conductor current so as to eliminate the lead error in the prediction process;
(3) the method and the system endow different weights to each historical moment value through a heuristic optimization algorithm and obtain optimized weight factors so as to accurately calculate the temperature of the cable conductor at the current moment;
(4) the method and the system are easy to realize, simple in model and quick in calculation, are suitable for any complex external environment and multi-loop cable situation, and have reference significance for on-line monitoring, state evaluation, capacity increase and efficiency improvement of the high-voltage power cable and the like.
Drawings
FIG. 1 is a flowchart of the steps for training a cable conductor temperature prediction model in accordance with a preferred embodiment of the present invention;
FIG. 2 is a flow chart of the steps of the method according to the preferred embodiment of the present invention;
FIG. 3 is a schematic diagram of a cable steady state thermal circuit model according to a preferred embodiment of the present invention;
FIG. 4 is a flowchart illustrating the detailed process of step A2 according to the preferred embodiment of the present invention;
FIG. 5 is a schematic diagram of the system in accordance with the preferred embodiment of the present invention;
FIG. 6 is a graph of conductor current under a long cycle (24h) load in a second embodiment in accordance with the present invention;
FIG. 7 is a graph of conductor current under short cycle (6h) load in the second embodiment of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present invention is not limited to the embodiment, and other embodiments may be included in the scope of the present invention as long as the gist of the present invention is satisfied.
In a preferred embodiment of the present invention, based on the above problems in the prior art, a cable conductor temperature prediction method considering historical data is provided, and a cable conductor temperature prediction model is obtained by pre-training, as shown in fig. 1, which specifically includes the following steps:
step A1, collecting current monitoring data, multiple groups of historical monitoring data and corresponding structural data of at least one high-voltage cable, wherein the current monitoring data comprises a skin temperature, a conductor current and a conductor temperature at the current data collection time, and each group of historical monitoring data comprises a cable skin temperature and a cable conductor current at the multiple historical data collection times;
a2, taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current and the structural data of each historical data acquisition moment as input, taking the conductor temperature as output, and training to obtain a cable conductor temperature prediction model;
the method for predicting the temperature of the cable conductor is shown in fig. 2, and specifically includes the following steps:
step S1, collecting current conductor monitoring data of a cable to be predicted at the current collecting moment, historical conductor monitoring data of at least one historical collecting moment and corresponding cable structure data;
step S2, inputting the current conductor monitoring data, the historical conductor monitoring data and the cable structure data into the cable conductor temperature prediction model to obtain a conductor temperature prediction value.
Specifically, in this embodiment, the cable skin temperature and the cable conductor current of the high-voltage cable over a period of time are collected by a field test or finite element simulation method as historical monitoring data required for training a cable conductor temperature prediction model.
Preferably, the collection time of the historical monitoring data may be several days (e.g. 5 days), the collection time interval between the collection times of the historical data may be several minutes (e.g. 5 minutes), and the historical monitoring data may be a section of recorded data of the high-voltage cable in any operating state.
Preferably, the current monitoring data and the multiple groups of historical monitoring data used in the model training are different from the current conductor monitoring data and the historical conductor monitoring data used in the model prediction, the current monitoring data and the multiple groups of historical monitoring data used in the model training are the cable skin temperature, the cable conductor current and the cable conductor temperature of the high-voltage cable within a period of time, and the cable conductor temperature at each moment within the period of time is known.
Preferably, the current monitoring data and the multiple groups of historical monitoring data used in the model training can be training data prepared in advance, and do not need to be acquired before the model training.
In a preferred embodiment of the present invention, the current conductor monitoring data includes a current cable skin temperature and a current cable conductor current at a current collection time, and the historical conductor monitoring data includes a historical cable skin temperature and a historical cable conductor current corresponding to at least one historical collection time before the current collection time.
In the preferred embodiment of the present invention, the structural data includes an insulation layer equivalent thermal resistance value, an outer sheath equivalent thermal resistance value, an insulation layer loss value and an outer sheath loss value of the high voltage cable.
Specifically, in the present embodiment, based on IEC60287 standard, according to the current situationEstablishing a cable steady-state thermal circuit model by monitoring data and structural data, wherein only thermal resistance is considered and thermal capacity is not considered in the cable steady-state thermal circuit model, the cable steady-state thermal circuit model is shown in figure 3, R1Is equivalent thermal resistance of insulating layer, R2Is equivalent thermal resistance of the outer sheath, Q represents total loss of the cable and is the sum of loss value of the conductor, loss value of the insulating layer and loss value of the outer sheath, TcIndicating the conductor temperature, TsThe skin temperature is indicated.
In the preferred embodiment of the present invention, as shown in fig. 4, step a2 includes:
a21, processing according to the cable skin temperature at each historical data acquisition time, the skin temperature at the current data acquisition time and a preset weight factor to obtain a skin temperature equivalent value, and processing according to the cable conductor current at each historical data acquisition time, the conductor current at the current data acquisition time and the weight factor to obtain a conductor current equivalent value;
a22, processing according to the skin temperature equivalent value, the conductor current equivalent value, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value, the outer sheath loss value and the weight factor to obtain a temperature prediction expression;
and A23, based on the temperature prediction expression, taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value and the outer sheath loss value of the current data acquisition time as input, taking the conductor temperature as output, and training to obtain a cable conductor temperature prediction model.
Specifically, in this embodiment, a relational expression is proposed according to the conductor temperature, the skin temperature, the total cable loss, the insulation layer equivalent thermal resistance value, and the outer sheath equivalent thermal resistance value before establishing the temperature prediction expression, and based on the hot spot analogy theory, the relational expression under the steady state condition is easily obtained as follows:
Tc＝Ts+Q(R1+R2)
wherein,
Tcrepresents the conductor temperature;
Tsindicating the skin temperature;
q represents the total loss of the cable;
R1representing the equivalent thermal resistance value of the insulating layer;
R2indicating the equivalent thermal resistance of the outer sheath.
Preferably, the cable total loss is calculated by the following calculation formula:
Q＝Ic 2Rα+Wd+Wj
wherein,
q represents the total loss of the cable;
Icrepresenting a conductor current;
Rαrepresenting the conductor resistance;
Wdrepresenting an insulation layer loss value;
Wjrepresenting the outer sheath loss value.
Preferably, considering that the change of the current of the conductor of the cable is rapid, but the response time of the thermal process of the cable is hours, the direct use of the relational expression to train the temperature prediction model of the conductor of the cable can cause a large lead error, so that the temperature prediction model is optimized by considering historical monitoring data.
Preferably, when calculating the conductor temperature at the current data acquisition time, the skin temperature equivalent value is calculated by taking into account the contribution of the historical time values of the two parameters, namely the skin temperature and the conductor current, to the conductor temperature by the following calculation formula:
wherein,
ts (eq) represents the skin temperature equivalent value;
wi represents a weight factor, Wi ≧ 0 and
Δ t represents a sampling time interval between adjacent historical data acquisition moments;
preferably, the conductor current equivalent value is calculated by the following calculation formula:
wherein,
ic (eq) represents the skin temperature equivalent value;
Wirepresents a weight factor, Wi ≧ 0 and
Δ t represents the sampling time interval between adjacent historical data acquisition instants.
Preferably, the skin temperature equivalent value is substituted for the skin temperature in the relational expression, and the conductor current equivalent value is substituted for the conductor current in the relational expression, so as to obtain the temperature prediction expression.
In a preferred embodiment of the present invention, before the step a21 is executed, the weighting factor is optimized according to a heuristic optimization algorithm to obtain an optimized weighting factor, then in the step a21, the skin temperature equivalent value is obtained by processing according to the cable skin temperature at each historical data acquisition time, the skin temperature at the current data acquisition time, and the optimized weighting factor, and the conductor current equivalent value is obtained by processing according to the cable conductor current at each historical data acquisition time, the conductor current at the current data acquisition time, and the optimized weighting factor.
Specifically, in this embodiment, the weighting factor is optimized through a heuristic optimization algorithm (such as a genetic algorithm, a simulated annealing algorithm, an ant colony algorithm, and the like), so that the total prediction error of the conductor temperature in the historical monitoring data is minimized, and the weighting factor to be optimized is W ═ W0,W1,W2,…,Wn]Wherein n is a natural number.
Preferably, in this embodiment, a genetic algorithm is adopted to optimize the weight factor, an evolutionary algebra counter and a maximum evolutionary algebra are set first, a plurality of individuals are randomly generated as an initial population, then the fitness of each individual in the population is calculated, a selection operator is acted on the population, on the basis of the fitness evaluation of the individuals in the population, a crossover operator and a mutation operator are sequentially acted on the population, the population is subjected to selection, crossover and mutation operations to obtain a next generation population, then an optimal solution is judged and output according to a termination condition, and the calculation is terminated to obtain the optimized weight factor.
Preferably, in the actual operation process, the cable conductor temperature prediction model may be trained according to the data of the first hour, the second hour and the third hour, at this time, an initial weight factor is obtained, then the cable conductor temperature prediction model is trained according to the data of the second hour, the third hour and the fourth hour, at this time, a weight factor after the first optimization is obtained, and the weight factor is continuously subjected to iterative optimization along with the increase of the training data and the training times.
In a preferred embodiment of the present invention, a set of historical monitoring data is collected in step a1, and then in step S1, the historical conductor monitoring data includes historical cable skin temperature and historical cable conductor current corresponding to a historical collection time before the current collection time.
Specifically, in this embodiment, if the current acquisition time is the 2 nd hour and the acquisition time interval is 1 hour, the historical acquisition time before the current acquisition time is the 1 st hour.
In a preferred embodiment of the present invention, in step a1, a plurality of sets of historical monitoring data are collected, and in step S1, the historical monitoring data of the conductor includes historical cable skin temperatures and historical cable conductor currents corresponding to a plurality of historical collection times before the current collection time.
Specifically, in this embodiment, if the current acquisition time is the 3 rd hour and the acquisition time interval is 1 hour, a plurality of historical acquisition times before the current acquisition time may be the 2 nd hour and the 1 st hour, respectively, and the number of the historical acquisition times may be adjusted correspondingly along with the change of the current acquisition time.
Specifically, in this embodiment, the conductor temperature predicted value obtained by collecting multiple sets of historical monitoring data is compared with the conductor temperature predicted value obtained by collecting only one set of historical monitoring data, and it is found through comparison that the conductor temperature predicted value obtained by collecting multiple sets of historical monitoring data is more accurate than the conductor temperature predicted value obtained by collecting only one set of historical monitoring data.
Preferably, the number of sets of historical monitoring data may be 2, 3 or more.
In a preferred embodiment of the present invention, a cable conductor temperature prediction system taking historical data into account is applied to the cable conductor temperature prediction method, as shown in fig. 5, and includes:
a model training module 1, comprising:
the data acquisition unit 11 is used for acquiring current monitoring data of at least one high-voltage cable, at least one group of historical monitoring data and corresponding structural data, wherein the current monitoring data comprises a skin temperature, a conductor current and a conductor temperature at the current data acquisition moment, and each group of historical monitoring data comprises a cable skin temperature and a cable conductor current at a plurality of historical data acquisition moments;
the model training unit 12 is connected with the data acquisition unit 11 and used for training to obtain a cable conductor temperature prediction model by taking the cable skin temperature, the cable conductor current, the skin temperature at the current data acquisition time, the conductor current and the structural data as input and the conductor temperature as output at each historical data acquisition time;
the data acquisition module 2 is used for acquiring current conductor monitoring data of a cable to be predicted at the current acquisition time, historical conductor monitoring data of at least one historical acquisition time and corresponding cable structure data;
and the temperature prediction module 3 is respectively connected with the data acquisition module 2 and the model training module 1 and is used for inputting the current conductor monitoring data, the historical conductor monitoring data and the cable structure data into the cable conductor temperature prediction model to obtain a conductor temperature prediction value.
In a preferred embodiment of the present invention, the structural data includes an insulation layer equivalent thermal resistance value, an outer sheath equivalent thermal resistance value, an insulation layer loss value and an outer sheath loss value of the high voltage cable, and the model training unit 12 includes:
the first processing subunit 121 is configured to obtain a skin temperature equivalent value by processing according to the skin temperature of the cable at each historical data acquisition time, the skin temperature at the current data acquisition time, and a preset weight factor, and obtain a conductor current equivalent value by processing according to the conductor current of the cable at each historical data acquisition time, the conductor current at the current data acquisition time, and the weight factor;
the second processing subunit 122 is connected with the first processing subunit 121 and is used for processing according to the skin temperature equivalent value, the conductor current equivalent value, the insulating layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulating layer loss value, the outer sheath loss value and the weight factor to obtain a temperature prediction expression;
and the model training subunit 123 is connected to the second processing subunit 122, and is configured to train to obtain the cable conductor temperature prediction model by taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current, the insulating layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulating layer loss value, and the outer sheath loss value at each historical data acquisition time as inputs and taking the conductor temperature as an output, based on the temperature prediction expression.
In a preferred embodiment of the present invention, the model training unit 12 further includes a factor optimization subunit 124, respectively connected to the first processing subunit 121 and the second processing subunit 122, for optimizing the weight factor according to a heuristic optimization algorithm to obtain an optimized weight factor and controlling the first processing subunit 121 and the second processing subunit 122 to update the weight factor to the optimized weight factor.
Specifically, the first embodiment:
the single-loop single-core crosslinked polyethylene cable laid in the groove is taken as an example for explanation, the cable model is 110kV-YJLW02, structural parameters and thermal parameters of the cable can be consulted with relevant manuals, a finite element model is established based on the structural parameters, the thermal parameters and the laying environment parameters of the cable, the skin temperature, the conductor current and the conductor temperature required by model optimization are obtained through finite element simulation, the simulation time period is set to be 168 hours, the acquisition time interval is 5 minutes, the current applied to the conductor in the simulation model is generated randomly, and the current value is updated every hour.
Preferably, in the first embodiment, the skin temperature, the conductor current and the conductor temperature of the cable for 168 hours are collected as training data of a cable conductor temperature prediction model, rather than current monitoring data and historical monitoring data required by the cable conductor temperature prediction model.
Preferably, in the actual training process, the skin temperature, the conductor current, the skin temperature and the conductor current measured at 1 hour and 2 hours of the cable 168 hours and measured at 3 hours can be used as input data, the conductor temperature measured at 3 hours can be used as output data to train the cable conductor temperature prediction model, and the training method can be performed according to the data at 2 hours, 3 hours and 4 hours, and the like.
Specifically, the second embodiment:
the method comprises the steps of considering a long-period (24h) load and a short-period (6h) load and adopting three data acquisition methods, wherein the historical acquisition time is not considered in the first data acquisition method, one historical acquisition time is considered in the second data acquisition method, and two historical acquisition times are considered in the third data acquisition method, wherein the conductor current change trend under the long-period (24h) load is shown in figure 6, the conductor current change trend under the short-period (6h) load is shown in figure 7, the horizontal axis in the figure is time, and the vertical axis in the figure is conductor current.
Preferably, the first data acquisition method is to directly input the conductor current and the skin temperature at the current acquisition time into the cable conductor temperature prediction model to obtain a conductor temperature prediction value at the current acquisition time, and the historical acquisition time is not considered in the method.
Preferably, the second data acquisition method is to input the conductor current and the skin temperature at the current acquisition time and the conductor current and the skin temperature at a historical acquisition time adjacent to the current acquisition time into the cable conductor temperature prediction model to obtain a conductor temperature predicted value at the current acquisition time, for example, the 2 nd hour is taken as the current acquisition time, and the 1 st hour is taken as the historical acquisition time.
Preferably, the third data acquisition method is to input the conductor current and the skin temperature at the current acquisition time and the conductor current and the skin temperature at two historical acquisition times adjacent to the current acquisition time into the cable conductor temperature prediction model to obtain a conductor temperature predicted value at the current acquisition time, for example, the 3 rd hour is taken as the current acquisition time, the 2 nd hour is taken as the first historical acquisition time, and the 1 st hour is taken as the second historical acquisition time.
Preferably, the mean square error for the first data acquisition method is calculated to be 3.75 ℃ under a long period (24h) load2The mean error is 1.60 ℃, the maximum error is 4.34 ℃, and the mean square error of the second data acquisition method is 0.84 DEG C2The mean error is 0.73 ℃, the maximum error is 2.23 ℃, and the mean square error of the third data acquisition method is 0.07 DEG C2The average error is 0.22 ℃ and the maximum error is 0.74 ℃.
Preferably, the mean square error of the first data acquisition method is calculated to be 14.73 ℃ under a short period (6h) load2The mean error is 3.37 ℃, the maximum error is 7.81 ℃, and the mean square error of the second data acquisition method is 1.86 DEG C2The mean error is 1.22 ℃, the maximum error is 2.69 ℃, and the mean square error of the third data acquisition method is 0.55 DEG C2The average error is 0.63 ℃ and the maximum error is 1.59 ℃.
Preferably, the error of the first data acquisition method is found to be the largest through calculation, and the error of the third data acquisition method is found to be the smallest through calculation, so that the increase of the number of the considered historical moments is helpful for reducing the error of the predicted value of the conductor temperature, and particularly when the load is in rapid change, the model prediction result can well track the change of the actual conductor temperature.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (10)
1. A cable conductor temperature prediction method considering historical data is characterized in that a cable conductor temperature prediction model is obtained through pre-training, and the method specifically comprises the following steps:
step A1, acquiring current monitoring data, at least one group of historical monitoring data and corresponding structural data of at least one high-voltage cable, wherein the current monitoring data comprises a skin temperature, a conductor current and a conductor temperature at the current data acquisition moment, and each group of historical monitoring data comprises a cable skin temperature and a cable conductor current at a plurality of historical data acquisition moments;
step A2, taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current and the structural data of each historical data acquisition moment as input, taking the conductor temperature as output, and training to obtain a cable conductor temperature prediction model;
the method for predicting the temperature of the cable conductor specifically comprises the following steps:
step S1, collecting current conductor monitoring data of a cable to be predicted at the current collecting moment, historical conductor monitoring data of at least one historical collecting moment and corresponding cable structure data;
and step S2, inputting the current conductor monitoring data, the historical conductor monitoring data and the cable structure data into the cable conductor temperature prediction model to obtain a conductor temperature prediction value.
2. The method of claim 1, wherein the current conductor monitoring data includes a current cable skin temperature and a current cable conductor current at a current collection time, and the historical conductor monitoring data includes a historical cable skin temperature and a historical cable conductor current corresponding to at least one historical collection time prior to the current collection time.
3. The method of claim 1, wherein the structural data includes an insulation layer equivalent thermal resistance value, an outer sheath equivalent thermal resistance value, an insulation layer loss value, and an outer sheath loss value of the high voltage cable.
4. The method for predicting the conductor temperature of the cable according to claim 3, wherein the step A2 comprises:
step A21, processing according to the skin temperature of the cable at each historical data acquisition time, the skin temperature at the current data acquisition time and a preset weight factor to obtain a skin temperature equivalent value, and processing according to the conductor current of the cable at each historical data acquisition time, the conductor current at the current data acquisition time and the weight factor to obtain a conductor current equivalent value;
step A22, processing according to the skin temperature equivalent value, the conductor current equivalent value, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value, the outer sheath loss value and the weight factor to obtain a temperature prediction expression;
step A23, based on the temperature prediction expression, taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value and the outer sheath loss value of each historical data acquisition moment as input, taking the conductor temperature as output, and training to obtain the cable conductor temperature prediction model.
5. The method as claimed in claim 4, wherein before the step A21 is executed, the weighting factors are optimized according to a heuristic optimization algorithm to obtain an optimized weighting factor, and then in the step A21, the skin temperature equivalent value is obtained by processing according to the cable skin temperature at each historical data acquisition time, the skin temperature at the current data acquisition time, and the optimized weighting factor, and the conductor current equivalent value is obtained by processing according to the cable conductor current at each historical data acquisition time, the conductor current at the current data acquisition time, and the optimized weighting factor.
6. The method of claim 2, wherein a set of historical monitoring data is collected in step A1, and the historical cable skin temperature and the historical cable conductor current corresponding to a historical collection time prior to the current collection time are included in the historical conductor monitoring data in step S1.
7. The method of claim 2, wherein a plurality of sets of historical monitoring data are collected in the step A1, and the historical cable skin temperature and the historical cable conductor current corresponding to a plurality of historical collection times before the current collection time are included in the historical conductor monitoring data in the step S1.
8. A cable conductor temperature prediction system taking historical data into account, which is applied to the real-time temperature prediction method according to any one of claims 1 to 7, and comprises:
a model training module, comprising:
the data acquisition unit is used for acquiring current monitoring data, at least one group of historical monitoring data and corresponding structural data of at least one high-voltage cable, wherein the current monitoring data comprises a skin temperature, a conductor current and a conductor temperature at the current data acquisition moment, and each group of historical monitoring data comprises a cable skin temperature and a cable conductor current at a plurality of historical data acquisition moments;
the model training unit is connected with the data acquisition unit and used for training the cable skin temperature, the cable conductor current, the skin temperature, the conductor current and the structural data at the current data acquisition time as input, and the conductor temperature as output to obtain the cable conductor temperature prediction model;
the data acquisition module is used for acquiring current conductor monitoring data of a cable to be predicted at the current acquisition moment, historical conductor monitoring data of at least one historical acquisition moment and corresponding cable structure data;
and the temperature prediction module is respectively connected with the data acquisition module and the model training module and is used for inputting the current conductor monitoring data, the historical conductor monitoring data and the cable structure data into the cable conductor temperature prediction model to obtain a conductor temperature prediction value.
9. The cable conductor temperature prediction system of claim 8, wherein the structural data includes an insulation layer equivalent thermal resistance value, an outer sheath equivalent thermal resistance value, an insulation layer loss value, and an outer sheath loss value of the high voltage cable, and the model training unit includes:
the first processing subunit is used for processing the cable skin temperature at each historical data acquisition moment, the skin temperature at the current data acquisition moment and a preset weight factor to obtain a skin temperature equivalent value, and processing the cable conductor current at each historical data acquisition moment, the conductor current at the current data acquisition moment and the weight factor to obtain a conductor current equivalent value;
the second processing subunit is connected with the first processing subunit and used for processing according to the skin temperature equivalent value, the conductor current equivalent value, the insulating layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulating layer loss value, the outer sheath loss value and the weight factor to obtain a temperature prediction expression;
and the model training subunit is connected with the second processing subunit and used for training to obtain the cable conductor temperature prediction model by taking the cable skin temperature, the cable conductor current, the skin temperature, the conductor current, the insulation layer equivalent thermal resistance value, the outer sheath equivalent thermal resistance value, the insulation layer loss value and the outer sheath loss value of each historical data acquisition moment as input and taking the conductor temperature as output based on the temperature prediction expression.
10. The system of claim 9, wherein the model training unit further comprises a factor optimization subunit, respectively connected to the first processing subunit and the second processing subunit, for optimizing the weighting factor according to a heuristic optimization algorithm to obtain an optimized weighting factor and controlling the first processing subunit and the second processing subunit to update the weighting factor to the optimized weighting factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111449826.2A CN114323337B (en) | 2021-11-30 | 2021-11-30 | Cable conductor temperature prediction method and system considering historical data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111449826.2A CN114323337B (en) | 2021-11-30 | 2021-11-30 | Cable conductor temperature prediction method and system considering historical data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114323337A true CN114323337A (en) | 2022-04-12 |
| CN114323337B CN114323337B (en) | 2024-09-17 |
Family
ID=81048005
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111449826.2A Active CN114323337B (en) | 2021-11-30 | 2021-11-30 | Cable conductor temperature prediction method and system considering historical data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114323337B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN116147805A (en) * | 2023-04-20 | 2023-05-23 | 北京工业大学 | Redundant monitoring method and system for monitoring temperature of stuffing box of pump station |
| CN117169658A (en) * | 2023-11-03 | 2023-12-05 | 四川省东舜智能科技有限公司 | Photoelectric hybrid cable monitoring method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005283148A (en) * | 2004-03-26 | 2005-10-13 | Tokyo Electric Power Co Inc:The | Conductor temperature estimation method of power cable |
| WO2010021287A1 (en) * | 2008-08-18 | 2010-02-25 | 日本電気株式会社 | Substrate warpage predicting method, substrate warpage predicting system, and substrate warpage predicting program |
| WO2017113875A1 (en) * | 2015-12-29 | 2017-07-06 | 广州视源电子科技股份有限公司 | Temperature prediction method and system thereof |
| CN112115636A (en) * | 2020-08-18 | 2020-12-22 | 国网山东省电力公司经济技术研究院 | Method and system for predicting insulation aging life of power cable in advance |
| CN113486573A (en) * | 2021-06-18 | 2021-10-08 | 国网江西省电力有限公司电力科学研究院 | Cable core temperature prediction method |
-
2021
- 2021-11-30 CN CN202111449826.2A patent/CN114323337B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005283148A (en) * | 2004-03-26 | 2005-10-13 | Tokyo Electric Power Co Inc:The | Conductor temperature estimation method of power cable |
| WO2010021287A1 (en) * | 2008-08-18 | 2010-02-25 | 日本電気株式会社 | Substrate warpage predicting method, substrate warpage predicting system, and substrate warpage predicting program |
| WO2017113875A1 (en) * | 2015-12-29 | 2017-07-06 | 广州视源电子科技股份有限公司 | Temperature prediction method and system thereof |
| CN112115636A (en) * | 2020-08-18 | 2020-12-22 | 国网山东省电力公司经济技术研究院 | Method and system for predicting insulation aging life of power cable in advance |
| CN113486573A (en) * | 2021-06-18 | 2021-10-08 | 国网江西省电力有限公司电力科学研究院 | Cable core temperature prediction method |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN116147805A (en) * | 2023-04-20 | 2023-05-23 | 北京工业大学 | Redundant monitoring method and system for monitoring temperature of stuffing box of pump station |
| CN117169658A (en) * | 2023-11-03 | 2023-12-05 | 四川省东舜智能科技有限公司 | Photoelectric hybrid cable monitoring method |
| CN117169658B (en) * | 2023-11-03 | 2024-02-02 | 四川省东舜智能科技有限公司 | Photoelectric hybrid cable monitoring method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114323337B (en) | 2024-09-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114323337B (en) | Cable conductor temperature prediction method and system considering historical data | |
| CN107957562B (en) | Online prediction method for residual life of lithium ion battery | |
| CN105137242B (en) | Single-phase photovoltaic inverter on-line condition monitoring and method for predicting residual useful life | |
| Illias et al. | Estimation of transformer parameters from nameplate data by imperialist competitive and gravitational search algorithms | |
| CN111669123B (en) | Method and device for fault diagnosis of photovoltaic string | |
| CN115952698B (en) | Method and system for optimizing thermal energy management of lithium battery energy storage box | |
| CN110969306A (en) | Power distribution low-voltage distribution area load prediction method and device based on deep learning | |
| CN117318111B (en) | Weather prediction-based dynamic adjustment method and system for light energy storage source | |
| CN112036067A (en) | Method for predicting steady-state temperature rise of groove cable | |
| Fu et al. | Modelling and prediction techniques for dynamic overhead line rating | |
| Stuchly et al. | A power quality forecasting model as an integrate part of active demand side management using Artificial Intelligence Technique-Multilayer Neural Network with Backpropagation Learning Algorithm | |
| CN115840119A (en) | Power cable line degradation diagnostic system and method using database samples | |
| CN118017576A (en) | Net-structured string-type energy storage converter | |
| CN115034493A (en) | Wind power plant black start path optimization method considering unit operation state | |
| CN114565182A (en) | Power equipment temperature prediction method based on non-dominated sorting genetic optimization bidirectional long-time memory neural network | |
| CN114091316A (en) | Regional photovoltaic power generation capacity calculation method | |
| CN117148048A (en) | Power distribution network fault prediction method and system based on digital twin technology | |
| CN116885853A (en) | Automatic power distribution monitoring system based on Internet of things | |
| Yang et al. | Overhead conductor thermal dynamics identification by using echo state networks | |
| Tina et al. | Thermal parameter identification of photovoltaic module using genetic algorithm | |
| CN109546966A (en) | A kind of the hot spot diagnostic method and device of photovoltaic module | |
| CN107742886B (en) | Prediction method for load peak simultaneous coefficient of thermoelectric combined system | |
| Tian et al. | Fusion modeling for wind power forecasting based on redundant method elimination | |
| Wei et al. | Frequency prediction of synchronous generators in a multi-machine power system with a photovoltaic plant using a cellular computational network | |
| CN117713211B (en) | Photovoltaic grid-connected intelligent scheduling method and system based on environmental analysis |
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 | ||
| GR01 | Patent grant |