CN108764550A - Lightning Warning method and system based on transmission line information data - Google Patents
Lightning Warning method and system based on transmission line information data Download PDFInfo
- Publication number
- CN108764550A CN108764550A CN201810481599.3A CN201810481599A CN108764550A CN 108764550 A CN108764550 A CN 108764550A CN 201810481599 A CN201810481599 A CN 201810481599A CN 108764550 A CN108764550 A CN 108764550A
- Authority
- CN
- China
- Prior art keywords
- weather
- parameter
- parameters
- time
- lightning
- 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
- 230000005540 biological transmission Effects 0.000 title claims abstract description 68
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000003062 neural network model Methods 0.000 claims abstract description 37
- 230000005611 electricity Effects 0.000 claims abstract description 32
- 238000001514 detection method Methods 0.000 claims abstract description 19
- 238000013528 artificial neural network Methods 0.000 claims abstract description 15
- 238000005286 illumination Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 21
- 238000004891 communication Methods 0.000 claims description 5
- 230000005672 electromagnetic field Effects 0.000 claims description 5
- 238000013500 data storage Methods 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 210000004218 nerve net Anatomy 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 8
- 241000208340 Araliaceae Species 0.000 description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 4
- 235000008434 ginseng Nutrition 0.000 description 4
- 238000002156 mixing Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Operations Research (AREA)
- Primary Health Care (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Biomedical Technology (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Telephonic Communication Services (AREA)
Abstract
The application provides a kind of Lightning Warning method and system based on transmission line information data, the method first obtains the real-time weather parameter being tested in transmission line faultlocating environment at current time, real-time weather parameter is inputted into weather data prediction model again, obtain the weather forecasting parameter of subsequent time, the weather conditions of subsequent time are finally determined according to the weather forecasting parameter, to monitor and predict the weather conditions on transmission line of electricity in real time.Wherein, the weather data prediction model is the BP neural network model generated according to weather history parameter, not only the weather conditions in detection environment can in real time be predicted by BP neural network, make the method for early warning that there is higher real-time, and it can be by constantly recording the weather parameters in detection environment, weather data prediction model is constantly corrected, the reliability of method for early warning is improved, solves the problems, such as traditional method for early warning real-time and poor reliability.
Description
Technical field
This application involves technical field of power systems more particularly to a kind of Lightning Warnings based on transmission line information data
Method and system.
Background technology
Growth with country to electricity consumption demand, high voltage, low-voltage transmission line of electricity also increasingly increase severely.China's topography is multiple
Miscellaneous changeable, mountains and rivers, river are numerous, and transmission line of electricity is often needed across a variety of different terrain environments, for example, coming for Yunnan Province
It says, transmission line of electricity is often passed through from river valley, intermountain, in these areas, is easy to form a variety of different localized weather environment,
Such as the bad weathers such as thunder and lightning usually concentrate transmission line of electricity, are impacted to transmission line of electricity.And since traffic is constant, in thunder and lightning
After hitting transmission line of electricity transmission of electricity being caused to be interrupted, it can not be found in time, to influence nearby resident and commercial power.
In order to mitigate influence of the weather condition to transmission line of electricity, in the prior art, the weather that general logical meteorological department provides
Forecast information makes strick precaution in advance, and the normal operation of transmission line of electricity is influenced to avoid thunder and lightning weather.But in above-mentioned power transmission line
In environment residing for road, influenced that different meteorological conditions can be formed in regional area by mountains and rivers, river, such as in overall day
Under conditions of gas is sunny, some River Valley Regions can form local rainfall or thunder and lightning weather, and this local thunder and lightning weather is nothing
Method is obtained by the weather forecast information of meteorological department, and therefore, above-mentioned method for early warning has that reliability is poor.
Part can also be monitored by meteorological measuring station by the way that weather monitoring station is arranged in the environment on the spot in the prior art
Data are reported when there is local thunder and lightning weather, power operation department are made to learn a day gas bar in time by the weather environment in region
Whether part causes influence to transmission line of electricity.But it, can since weather monitoring station can only be after the thunder and lightning weather of part occurs
Weather data is reported, it is easy to before operation department does not make the precautionary measures, transmission line of electricity just by lightening strikes,
The normal operation of transmission line of electricity is also affected, i.e. the method for early warning real-time based on weather monitoring station is poor.
Invention content
This application provides a kind of Lightning Warning method and system based on transmission line information data, pre- to solve tradition
The problem of alarm method real-time and poor reliability.
On the one hand, the application provides a kind of Lightning Warning method based on transmission line information data, including:
The real-time weather parameter in current time tested transmission line faultlocating environment is obtained, the weather parameters includes temperature
Degree, humidity, intensity of illumination and Electromagnetic Signal Strength;
The real-time weather parameter is inputted into weather data prediction model, obtains the weather forecasting parameter of subsequent time;Institute
It is the BP neural network model generated according to weather history parameter to state weather data prediction model;The weather forecasting parameter includes
Temperature corresponding with the weather parameters, humidity, intensity of illumination and Electromagnetic Signal Strength predicted value;
The weather conditions of subsequent time are determined according to the weather forecasting parameter.
Optionally, the method is in the step for obtaining the real-time weather parameter that current time is tested in transmission line faultlocating environment
Before rapid, further include:
The weather history parameter being tested in transmission line faultlocating environment is obtained, the weather history parameter includes multiple moment
Temperature, humidity, intensity of illumination and the Electromagnetic Signal Strength of record;
It is input with the weather history parameter, the BP neural network model is trained, the day destiny is generated
It is predicted that model;
Store the weather data prediction model.
Optionally, it is input with the weather history parameter, the BP neural network model is trained, described in generation
The step of weather data prediction model, including:
The weather parameters of t moment is obtained from the weather history parameter;
The weather parameters of the t moment is inputted into the BP neural network model, obtains the weather parameters prediction at t+1 moment
Value;
Prediction error value is generated according to the weather parameters predicted value at the t+1 moment and the weather parameters value at t+1 moment;
Compare the prediction error value and preset error threshold;
If the prediction error value is less than or equal to the error threshold, determine that the BP neural network model is described
Weather data prediction model.
Optionally, it is input with the weather history parameter, the BP neural network model is trained, described in generation
The step of weather data prediction model, further include:
If the prediction error value is more than the error threshold, described in the BP neural network model backpropagation
The weather parameters at t+1 moment carries out right value update to the BP neural network model;
The weather parameters at t-1 moment is obtained from the weather history parameter;
It is input with the weather parameters at the t-1 moment, t is obtained by the BP neural network model after right value update
The weather parameters predicted value at moment;
Prediction error judgment is generated according to the weather parameters value of the weather parameters predicted value of the t moment and the t moment
Value, and the comparison prediction error judgment value and the error threshold;
If the prediction error judgment value is less than or equal to the error threshold, the god of the BP after right value update is determined
It is the weather data prediction model through network model;
If the prediction error judgment value is more than the error threshold, t- is obtained from the weather history parameter successively
The weather parameters at moment before 2 and t-2, and the prediction error judgment value is generated, until prediction error judgment value is small
In the prediction error judgment value.
Optionally, the weather parameters further includes:Observation area, observation unit where tested transmission line faultlocating environment
And the weather characteristics of tested transmission line faultlocating environment.
Optionally, before the real-time weather parameter being inputted weather data prediction model, the method further includes:
The real-time weather parameter is normalized according to the following formula;
Conversion value M1=(M-Min)/(Max-Min) of the weather parameters items;
In formula, M- initial values, Max- has recorded the maximum value of respective items in weather parameters, and Min- has been recorded in weather parameters
The minimum value of respective items.
Optionally, the step of weather conditions of subsequent time being determined according to the weather forecasting parameter, including:
Weather conditions by the subsequent time include on the screen of detection device;Alternatively,
By the weather conditions of the subsequent time by communication device, it is sent to and the tested transmission line faultlocating environment
Apart from nearest substation.
Optionally, the step of determining the weather conditions of subsequent time according to the weather forecasting parameter further include:
If the weather conditions of the subsequent time are thunder and lightning weather, thunder and lightning signal is generated;
The thunder and lightning signal is amplified, and drive voltage signal is generated according to the thunder and lightning signal and is sent to early warning dress
It sets, generates Lightning Warning.
On the other hand, the application also provides a kind of Lightning Warning system based on transmission line information data, including installation
Weather parameters sensor on being detected electric power line pole tower, and the data processing of the connection weather parameters sensor fill
It sets;Wherein:
The weather parameters sensor includes temperature sensor, humidity sensor, intensity of illumination sensor and electromagnetic field
Signal strength sensors;BP neural network computing unit and data storage disk are equipped in the data processing equipment;The data
Processing unit is further configured to execute following procedure step:
The real-time weather parameter in current time tested transmission line faultlocating environment is obtained, the weather parameters includes temperature
Degree, humidity, intensity of illumination and Electromagnetic Signal Strength;
The real-time weather parameter is inputted into weather data prediction model, obtains the weather forecasting parameter of subsequent time;Institute
It is the BP neural network model generated according to weather history parameter to state weather data prediction model;The weather forecasting parameter includes
Temperature corresponding with the weather parameters, humidity, intensity of illumination and Electromagnetic Signal Strength predicted value;
The weather conditions of subsequent time are determined according to the weather forecasting parameter.
Optionally, the system also includes the prior-warning devices for connecting the data processing equipment;The data processing equipment
It is further configured to execute following procedure step:
If the weather conditions of the subsequent time are thunder and lightning weather, thunder and lightning signal is generated;
The thunder and lightning signal is amplified, and drive voltage signal is generated according to the thunder and lightning signal and is sent to early warning dress
It sets, generates Lightning Warning.
By above technical scheme it is found that the application provide a kind of Lightning Warning method based on transmission line information data and
System, the method first obtain the real-time weather parameter being tested in transmission line faultlocating environment at current time, then by real-time weather
Parameter inputs weather data prediction model, the weather forecasting parameter of subsequent time is obtained, finally according to the weather forecasting parameter
The weather conditions of subsequent time are determined, to monitor and predict the weather conditions on transmission line of electricity in real time.Wherein, the day destiny
It is predicted that model is the BP neural network model generated according to weather history parameter, it not only can be by BP neural network to detection
Weather conditions in environment are predicted in real time, so that the method for early warning is had higher real-time, and can be by continuous
Weather parameters in record detection environment, constantly corrects weather data prediction model, improves the reliable of method for early warning
Property, solve the problems, such as traditional method for early warning real-time and poor reliability.
Description of the drawings
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without creative efforts, also
It can be obtain other attached drawings according to these attached drawings.
Fig. 1 is a kind of structural schematic diagram of BP neural network model;
Fig. 2 is a kind of flow diagram of the Lightning Warning method based on transmission line information data;
Fig. 3 is the flow diagram for generating weather data prediction model in the embodiment of the present application by weather parameters;
Fig. 4 is the flow diagram that weather data prediction model is determined in the embodiment of the present application;
Fig. 5 is the flow diagram that weather data prediction model is improved in the embodiment of the present application;
Fig. 6 is a kind of structural schematic diagram of the Lightning Warning system based on transmission line information data.
Specific implementation mode
Embodiment will be illustrated in detail below, the example is illustrated in the accompanying drawings.In the following description when referring to the accompanying drawings,
Unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Implementation described in following embodiment
Mode does not represent all embodiments consistent with the application.Only it is and be described in detail in claims, the application
The example of the consistent system and method for some aspects.
In technical solution provided by the present application, the transmission line information data refer to the environment residing for transmission line of electricity
In various types of environmental parameters, these parameters are the intrinsic environmental parameter determined due to orographic factor, this intrinsic ginseng a bit
Number is will not to change with the change of transmission line of electricity operating status, such as temperature, humidity, intensity of illumination in environment.And
Environmental parameter described herein also has one is can generate variation according to power line condition, the ginseng of this variation
Number can not only assistant analysis transmission line of electricity weather conditions, and the Changes in weather in surrounding subenvironment can be influenced indirectly,
Such as electromagnetic field intensity etc., although large effect can not be generated to environment, due to the presence of electromagnetic field, transmission line of electricity is more
It is easy to be hit by thunder and lightning so that the thunder and lightning weather for not interfering with transmission line of electricity theoretically also influences whether transmission line of electricity in practice
Operation.
Since the situation of change that these parameters can be used for reacting weather in this application will influence to transmit electricity
The parameter of weather, referred to as weather parameters in circuit ambient enviroment.In order to analyze and predict that conveniently, weather parameters includes in the application
Environmental parameter temperature, humidity, intensity of illumination and operating parameter electromagnetic field intensity, in addition, it can include for assistant analysis and
Position the ginsengs such as the observation area of transmission line of electricity local environment, the weather characteristics of observation unit and tested transmission line faultlocating environment
Number.
In technical solution provided by the present application, BP (back propagation) neural network model is a kind of artificial intelligence
It can model and a kind of multilayer feedforward neural network that can be trained according to error backpropagation algorithm.As shown in Figure 1, being one
The structural schematic diagram of kind BP neural network model.In BP neural network, including input layer, hidden layer and output layer, actually make
It can be inputted from input layer with middle training data, via the calculating of each layer, exported from output layer as a result, and can use
Corresponding verify data calculates the weights updated in each layer from output layer backpropagation, finally obtains the nerve for meeting business scenario
Network model.
It is a kind of flow diagram of the Lightning Warning method based on transmission line information data referring to Fig. 2, it can by Fig. 2
Know, Lightning Warning method provided by the present application includes the following steps:
S1:The real-time weather parameter in current time tested transmission line faultlocating environment is obtained, the weather parameters includes
Temperature, humidity, intensity of illumination and Electromagnetic Signal Strength;
S2:The real-time weather parameter is inputted into weather data prediction model, obtains the weather forecasting parameter of subsequent time;
The weather data prediction model is the BP neural network model generated according to weather history parameter;The weather forecasting parameter packet
Include temperature corresponding with the weather parameters, humidity, intensity of illumination and Electromagnetic Signal Strength predicted value;
S3:The weather conditions of subsequent time are determined according to the weather forecasting parameter.
In actual use, the real-time weather parameter for first passing through weather parameters sensor detection current time, due to transmission of electricity
The span of circuit may be very big, and the weather conditions difference of different regions is larger, therefore in practical application, can be to transmission line of electricity
It is divided into multiple detection zones in advance, corresponding weather parameters sensor is set in each detection zone.And in order to obtain
More accurate real-time weather parameter is obtained, real sensor may be mounted at close to the position of power transmission line, such as on the top of shaft tower
End, impacts the detection of sensor to avoid surrounding plants or river.
After sensor detects real-time weather parameter, the weather parameters detected is sent in data processing equipment,
In practical application, data processing equipment can be set along in the detection zone of transmission line of electricity with weather parameters sensor,
It can also be arranged in unified Central Control Room, it, can be with after detecting real-time weather parameter by weather parameters sensor at this time
The data detected are sent in the data processing equipment in Central Control Room by communication device, to predict the day of subsequent time
Vaporous condition and Lightning Warning.
It should be noted that in technical solution provided by the present application, the weather conditions refer to certain period of time with
Interior Changes in weather situation, for example, the Meteorological Characteristics such as sunny, cloudy, rainfall, snowfall, high wind, thunder and lightning.Due to actually using
In, thunder and lightning weather is affected for transmission line of electricity, therefore can simply set weather conditions to, has thunder and lightning and without thunder
Two kinds of electricity, or it is divided into different grades further according to the intensity of thunder and lightning weather, such as 0 grade without thunder and lightning, 1 grade of slight thunder and lightning, 2
The common thunder and lightning ... of grade
In the present embodiment, current time and subsequent time are pre-set detection time section, current time with it is next
Time interval between moment can be set according to the running environment of practical transmission line of electricity.For example, for climate variability
It coastal area can be by the smaller of time interval setting, i.e., since Changes in weather in different time periods has prodigious difference
More frequently weather conditions are predicted relatively, the generation for reducing burst thunder and lightning weather condition impacts prediction result;
And for the hinterland that weather is surely put, then it can suitably increase preset time interval, to reduce data processing amount, reduction pair
The configuration needs of data processing equipment.
In data processing equipment, it is correspondingly provided with BP neural network computing unit, in input real-time weather parameter
When, the weather parameters predicted value of subsequent time is exported, and according to the weather parameter data constantly obtained, predict weather data
Model is further trained, and the operation weights in continuous correction model make model be more in line with the day in current detection region
Gas feature.Therefore, as shown in figure 3, before carrying out weather parameters prediction, the application is further comprising the steps of, have been established with it is complete
It is apt to the weather data prediction model.
S21:The weather history parameter being tested in transmission line faultlocating environment is obtained, the weather history parameter includes multiple
Temperature, humidity, intensity of illumination and the Electromagnetic Signal Strength of moment record;
S22:It is input with the weather history parameter, the BP neural network model is trained, the day is generated
Gas data prediction model;
S23:Store the weather data prediction model.
Further, it is input with the weather history parameter, to the BP nerve nets as shown in figure 4, in step S22
Network model is trained, and is generated in the weather data prediction model, including:
S2201:The weather parameters of t moment is obtained from the weather history parameter;
S2202:The weather parameters of the t moment is inputted into the BP neural network model, obtains the weather ginseng at t+1 moment
Number predicted value;
S2203:Prediction is generated according to the weather parameters predicted value at the t+1 moment and the weather parameters value at t+1 moment to miss
Difference;
S2204:Compare the prediction error value and preset error threshold;
S2205:If the prediction error value is less than or equal to the error threshold, the BP neural network model is determined
For the weather data prediction model;
And in order to further improve the weather data prediction model, as shown in figure 5, being with the weather history parameter
In the step of inputting, being trained to the BP neural network model, generate the weather data prediction model, further include:
S2206:If the prediction error value is more than the error threshold, reversely passed by the BP neural network model
The weather parameters for broadcasting the t+1 moment carries out right value update to the BP neural network model;
S2207:The weather parameters at t-1 moment is obtained from the weather history parameter;
S2208:It is input with the weather parameters at the t-1 moment, passes through the BP neural network mould after right value update
Type obtains the weather parameters predicted value of t moment;
S2209:Prediction is generated according to the weather parameters value of the weather parameters predicted value of the t moment and the t moment to miss
Poor judgment value, and the comparison prediction error judgment value and the error threshold;
S2210:If the prediction error judgment value is less than or equal to the error threshold, the institute after right value update is determined
It is the weather data prediction model to state BP neural network model;
S2211:If the prediction error judgment value is more than the error threshold, successively from the weather history parameter
The weather parameters at moment before acquisition t-2 and t-2, and the prediction error judgment value is generated, until the prediction error is sentenced
Disconnected value is less than the prediction error judgment value.
In above-mentioned steps, moment t, t+1, t-1, t-2 etc., the time interval between represented adjacent moment should work as with above-mentioned
Time interval between preceding moment and subsequent time is consistent, i.e., "+1 " is only indicated with the number in " -1 " in the present embodiment
The quantity of corresponding time interval, does not indicate that specific time numerical value.
In the present embodiment, the weather parameters sensor detects 4 main weather parameters, i.e. temperature, humidity, illumination is strong
Degree and Electromagnetic Signal Strength, respective sensor detection data is a, b, c, d, to pass through BP neural network model prediction t+
The corresponding numerical value of weather parameters at 1 moment.Specifically, BP neural network blends to 4 parameters that sensor inputs, provide
One predicted value, then predicted value is compared analysis with actual value, if error is more than initial set value, continue input 4
Weather parameters, until predicted value and actual value are less than initial set value, BP god's network calculations are completed, due to after error judgment,
Be by the corresponding practical Value Data of output layer backpropagation, therefore multiple direction propagate after, the net of each layer in BP god's networks
Network weights are just determined, and model foundation is completed.Existing data, i.e. weather history parameter number may be used in the calculating of model weights
According to progress, calculating repeatedly improves precision of prediction for changing weights.
In practical calculating, BP god network is trained the weather parameters of input, is set into network topology initial weight,
Calculating continues, and provides primary network station result of calculation.Two parameter blendings, three parameters may be used in the blending of primary network station
Blending and four parameter blendings, output result can reach 4 layers or more as the input value of two grade network, the network number of plies,
Prediction result is better.But the number of plies of network is also unsuitable excessively high, because the excessively high network number of plies, not only increases operand, Er Qierong
Easily there is overcoupling phenomenon, reduces the precision of prediction of model instead.
Illustratively, under BP neural network mathematical model, input has following relationship with output:
Wherein:θ is threshold value, x=(x1..., xm)TInput vector is represented, y is output, wiIt is weight coefficient;F (X) is excitation
Function;It can be linear function, can also be nonlinear function;
In addition,
And can be in the signal of corresponding input layer and output layer:Input domain (input layer): Wherein:Vi (0) is to receive signal xi.
Domain output (output layer):Wherein:Vi (L) output signals yi.
For example, in actually calculating, BP neural network first carries out level one data fusion, level one data to weather parameter data
Fusion generates 9 secondary datas, then using secondary data as input value, is merged to secondary data, sets weights, and export
Lightning Warning value.The thunder and lightning predicted value and actual value of output are compared, when error is less than 4%, it is believed that computation model is reliable
And calculated repeatedly, weights are obtained, model foundation is completed.When error is more than 4%, model accuracy is inadequate, needs to data
Analysis.After the completion of BP neural network calculating, best a weights and threshold values can be obtained, i.e., best model foundation is complete
At, and can be as the computing unit of detection.BP neural network can be also modified at runtime, improve early warning precision.
In a kind of technical solution, step S2 is described before the real-time weather parameter is inputted weather data prediction model
Method further includes:
The real-time weather parameter is normalized according to the following formula;
Conversion value M1=(M-Min)/(Max-Min) of the weather parameters items;
In formula, M- initial values, Max- has recorded the maximum value of respective items in weather parameters, and Min- has been recorded in weather parameters
The minimum value of respective items.
The data of sensor input are normalized, can make the input values of all samples all close to 0 or
Person's numerical value smaller compared to mean square deviation, convenient for being formed with the input vector for being conducive to BP neural network model and calculating.
In the section Example of the application, S3 determines the weather conditions of subsequent time according to the weather forecasting parameter
The step of, including:
Weather conditions by the subsequent time include on the screen of detection device;Alternatively,
By the weather conditions of the subsequent time by communication device, it is sent to and the tested transmission line faultlocating environment
Apart from nearest substation.
Display screen accordingly can be set in actual use, in detection device, mould is predicted when passing through the weather data
After type acquires the data prediction of subsequent time, weather conditions are determined according to predicted value, and will determine that weather conditions result is shown
Show in screen, specific display mode can be shown by corresponding weather pattern figure, can also be not previously predicted
When to thunder and lightning weather, corresponding weather conditions are not shown, and when predicting subsequent time and will have thunder and lightning weather, with eye-catching
Color is shown.In the present embodiment, by communication device, the judging result of weather conditions is sent to and the tested transmission of electricity
Wireline inspection environment apart from nearest substation, can remind apart from nearest substation for subsequent time weather conditions and
When make corresponding Forewarning Measures.
In a kind of technical solution, S3, the step of weather conditions of subsequent time are determined according to the weather forecasting parameter,
Further include:
If the weather conditions of the subsequent time are thunder and lightning weather, thunder and lightning signal is generated;
The thunder and lightning signal is amplified, and drive voltage signal is generated according to the thunder and lightning signal and is sent to early warning dress
It sets, generates Lightning Warning.
Corresponding warning information can be generated by lightning warning device, after predicting thunder and lightning weather, notice is related
Unit or personnel carry out strick precaution early warning in advance.The prior-warning device, can be can generate any device of watchful signal, such as
Horn device, information platform etc..
Based on above-mentioned Lightning Warning, the application also provides a kind of Lightning Warning system based on transmission line information data,
As described in Figure 6, the early warning system includes being mounted on the weather parameters sensor 1, Yi Jilian being detected on electric power line pole tower
Connect the data processing equipment 2 of the weather parameters sensor 1;Wherein:
The weather parameters sensor include temperature sensor 11, humidity sensor 12, intensity of illumination sensor 13 and
Electromagnetic Signal Strength sensor 14;It is equipped with BP neural network computing unit 21 in the data processing equipment 2 and data store
Disk 22;The data processing equipment 2 is further configured to execute following procedure step:
The real-time weather parameter in current time tested transmission line faultlocating environment is obtained, the weather parameters includes temperature
Degree, humidity, intensity of illumination and Electromagnetic Signal Strength;
The real-time weather parameter is inputted into weather data prediction model, obtains the weather forecasting parameter of subsequent time;Institute
It is the BP neural network model generated according to weather history parameter to state weather data prediction model;The weather forecasting parameter includes
Temperature corresponding with the weather parameters, humidity, intensity of illumination and Electromagnetic Signal Strength predicted value;
The weather conditions of subsequent time are determined according to the weather forecasting parameter.
It should be noted that data storage disk 22 can not only be used for storing the weather data prediction model, it can be with
For storing the weather parameter data that weather parameters sensor 1 is detected, also, in order to avoid outdated data is to model accuracy
It influences, data storage disk 22 can periodically be reset.
Further, the system also includes the prior-warning devices 3 for connecting the data processing equipment 2;The data processing
Device 2 is further configured to execute following procedure step:
If the weather conditions of the subsequent time are thunder and lightning weather, thunder and lightning signal is generated;
The thunder and lightning signal is amplified, and drive voltage signal is generated according to the thunder and lightning signal and is sent to early warning dress
3 are set, Lightning Warning is generated.
By above technical scheme it is found that the application provide a kind of Lightning Warning method based on transmission line information data and
System, the method first obtain the real-time weather parameter being tested in transmission line faultlocating environment at current time, then by real-time weather
Parameter inputs weather data prediction model, the weather forecasting parameter of subsequent time is obtained, finally according to the weather forecasting parameter
The weather conditions of subsequent time are determined, to monitor and predict the weather conditions on transmission line of electricity in real time.Wherein, the day destiny
It is predicted that model is the BP neural network model generated according to weather history parameter, it not only can be by BP neural network to detection
Weather conditions in environment are predicted in real time, so that the method for early warning is had higher real-time, and can be by continuous
Weather parameters in record detection environment, constantly corrects weather data prediction model, improves the reliable of method for early warning
Property, solve the problems, such as traditional method for early warning real-time and poor reliability.
Similar portion cross-reference between embodiment provided by the present application, specific implementation mode provided above is only
It is several examples under the total design of the application, does not constitute the restriction of the application protection domain.For those skilled in the art
For member, any other embodiment expanded without creative efforts according to application scheme all belongs to
In the protection domain of the application.
Claims (10)
1. a kind of Lightning Warning method based on transmission line information data, which is characterized in that including:
Obtain the real-time weather parameter that current time is tested in transmission line faultlocating environment, the weather parameters includes temperature, wet
Degree, intensity of illumination and Electromagnetic Signal Strength;
The real-time weather parameter is inputted into weather data prediction model, obtains the weather forecasting parameter of subsequent time;The day
Gas data prediction model is the BP neural network model generated according to weather history parameter;The weather forecasting parameter includes and institute
State the corresponding temperature of weather parameters, humidity, intensity of illumination and Electromagnetic Signal Strength predicted value;
The weather conditions of subsequent time are determined according to the weather forecasting parameter, and record the real-time weather parameter.
2. Lightning Warning method according to claim 1, which is characterized in that the method is tested defeated at acquisition current time
Before electric line detects the step of real-time weather parameter in environment, further include:
The weather history parameter being tested in transmission line faultlocating environment is obtained, the weather history parameter includes multiple moment records
Temperature, humidity, intensity of illumination and Electromagnetic Signal Strength;
It is input with the weather history parameter, the BP neural network model is trained, it is pre- to generate the weather data
Survey model;
Store the weather data prediction model.
3. Lightning Warning method according to claim 2, which is characterized in that right with the weather history parameter to input
The step of BP neural network model is trained, generates the weather data prediction model, including:
The weather parameters of t moment is obtained from the weather history parameter;
The weather parameters of the t moment is inputted into the BP neural network model, obtains the weather parameters predicted value at t+1 moment;
Prediction error value is generated according to the weather parameters predicted value at the t+1 moment and the weather parameters value at t+1 moment;
Compare the prediction error value and preset error threshold;
If the prediction error value is less than or equal to the error threshold, determine that the BP neural network model is the weather
Data prediction model.
4. Lightning Warning method according to claim 3, which is characterized in that right with the weather history parameter to input
The step of BP neural network model is trained, generates the weather data prediction model further include:
If the prediction error value is more than the error threshold, pass through t+1 described in the BP neural network model backpropagation
The weather parameters at moment carries out right value update to the BP neural network model;
The weather parameters at t-1 moment is obtained from the weather history parameter;
It is input with the weather parameters at the t-1 moment, t moment is obtained by the BP neural network model after right value update
Weather parameters predicted value;
Prediction error judgment value is generated according to the weather parameters value of the weather parameters predicted value of the t moment and the t moment, with
And compare the prediction error judgment value and the error threshold;
If the prediction error judgment value is less than or equal to the error threshold, the BP nerve nets after right value update are determined
Network model is the weather data prediction model;
If the prediction error judgment value be more than the error threshold, successively from the weather history parameter obtain t-2 with
And before t-2 the moment weather parameters, and the prediction error judgment value is generated, until the prediction error judgment value is less than institute
State prediction error judgment value.
5. according to claim 1-4 any one of them Lightning Warning methods, which is characterized in that the weather parameters further includes:
The weather of observation area, observation unit and tested transmission line faultlocating environment where tested transmission line faultlocating environment is special
Sign.
6. Lightning Warning method according to claim 1, which is characterized in that the real-time weather parameter is inputted day destiny
It is predicted that before model, the method further includes:
The real-time weather parameter is normalized according to the following formula;
Conversion value M1=(M-Min)/(Max-Min) of the weather parameters items;
In formula, M- initial values, Max- has recorded the maximum value of respective items in weather parameters, and Min- has recorded corresponding in weather parameters
The minimum value of item.
7. Lightning Warning method according to claim 1, which is characterized in that determined according to the weather forecasting parameter next
The step of weather conditions at moment, including:
Weather conditions by the subsequent time include on the screen of detection device;Alternatively,
By the weather conditions of the subsequent time by communication device, it is sent to and the tested transmission line faultlocating environment distance
Nearest substation.
8. Lightning Warning method according to claim 1, which is characterized in that determined according to the weather forecasting parameter next
The step of weather conditions at moment, further include:
If the weather conditions of the subsequent time are thunder and lightning weather, thunder and lightning signal is generated;
The thunder and lightning signal is amplified, and drive voltage signal is generated according to the thunder and lightning signal and is sent to prior-warning device,
Generate Lightning Warning.
9. a kind of Lightning Warning system based on transmission line information data, which is characterized in that including being mounted on detected transmission of electricity
Weather parameters sensor on overhead line structures, and connect the data processing equipment of the weather parameters sensor;Wherein:
The weather parameters sensor includes temperature sensor, humidity sensor, intensity of illumination sensor and electromagnetic field signal
Intensity sensor;BP neural network computing unit and data storage disk are equipped in the data processing equipment;The data processing
Device is further configured to execute following procedure step:
Obtain the real-time weather parameter that current time is tested in transmission line faultlocating environment, the weather parameters includes temperature, wet
Degree, intensity of illumination and Electromagnetic Signal Strength;
The real-time weather parameter is inputted into weather data prediction model, obtains the weather forecasting parameter of subsequent time;The day
Gas data prediction model is the BP neural network model generated according to weather history parameter;The weather forecasting parameter includes and institute
State the corresponding temperature of weather parameters, humidity, intensity of illumination and Electromagnetic Signal Strength predicted value;
The weather conditions of subsequent time are determined according to the weather forecasting parameter.
10. Lightning Warning system according to claim 9, which is characterized in that the system also includes connect the data
The prior-warning device of processing unit;The data processing equipment is further configured to execute following procedure step:
If the weather conditions of the subsequent time are thunder and lightning weather, thunder and lightning signal is generated;
The thunder and lightning signal is amplified, and drive voltage signal is generated according to the thunder and lightning signal and is sent to prior-warning device,
Generate Lightning Warning.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810481599.3A CN108764550A (en) | 2018-05-18 | 2018-05-18 | Lightning Warning method and system based on transmission line information data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810481599.3A CN108764550A (en) | 2018-05-18 | 2018-05-18 | Lightning Warning method and system based on transmission line information data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108764550A true CN108764550A (en) | 2018-11-06 |
Family
ID=64008460
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810481599.3A Pending CN108764550A (en) | 2018-05-18 | 2018-05-18 | Lightning Warning method and system based on transmission line information data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108764550A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109886396A (en) * | 2019-03-18 | 2019-06-14 | 国家电网有限公司 | A kind of transmission line galloping on-line prediction system and method |
CN110160585A (en) * | 2019-05-24 | 2019-08-23 | 北京立群建筑科学研究院 | Environment detection method, device, equipment and storage medium |
CN111860997A (en) * | 2020-07-17 | 2020-10-30 | 海南大学 | Cross-data, information, knowledge modality and dimension early warning method and component |
CN113011098A (en) * | 2021-03-26 | 2021-06-22 | 云南电网有限责任公司电力科学研究院 | Power transmission corridor area lightning activity rule analysis display model and system |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002067013A2 (en) * | 2001-02-20 | 2002-08-29 | Massachusetts Institute Of Technology | Method and apparatus for short-term prediction of convective weather |
CN102590651A (en) * | 2011-12-31 | 2012-07-18 | 国网电力科学研究院 | Measured lightning data-based transmission line failure probability evaluation method |
CN103207340A (en) * | 2013-05-02 | 2013-07-17 | 深圳供电局有限公司 | On-line transmission line lightning shielding failure trip early-warning method |
CN106017551A (en) * | 2016-05-16 | 2016-10-12 | 国网河南省电力公司电力科学研究院 | Intelligent transmission line integrated monitoring analysis and early warning method |
CN106803129A (en) * | 2015-11-26 | 2017-06-06 | 中国电力科学研究院 | A kind of wind power ensemble prediction method based on multi-source numerical weather forecast |
CN107169598A (en) * | 2017-05-05 | 2017-09-15 | 上海斐讯数据通信技术有限公司 | A kind of weather prediction method and system based on deep learning |
-
2018
- 2018-05-18 CN CN201810481599.3A patent/CN108764550A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2002067013A2 (en) * | 2001-02-20 | 2002-08-29 | Massachusetts Institute Of Technology | Method and apparatus for short-term prediction of convective weather |
CN102590651A (en) * | 2011-12-31 | 2012-07-18 | 国网电力科学研究院 | Measured lightning data-based transmission line failure probability evaluation method |
CN103207340A (en) * | 2013-05-02 | 2013-07-17 | 深圳供电局有限公司 | On-line transmission line lightning shielding failure trip early-warning method |
CN106803129A (en) * | 2015-11-26 | 2017-06-06 | 中国电力科学研究院 | A kind of wind power ensemble prediction method based on multi-source numerical weather forecast |
CN106017551A (en) * | 2016-05-16 | 2016-10-12 | 国网河南省电力公司电力科学研究院 | Intelligent transmission line integrated monitoring analysis and early warning method |
CN107169598A (en) * | 2017-05-05 | 2017-09-15 | 上海斐讯数据通信技术有限公司 | A kind of weather prediction method and system based on deep learning |
Non-Patent Citations (1)
Title |
---|
刘白林主编: "《人工智能与专家***》", 28 February 2012, 西安交通大学出版社 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109886396A (en) * | 2019-03-18 | 2019-06-14 | 国家电网有限公司 | A kind of transmission line galloping on-line prediction system and method |
CN110160585A (en) * | 2019-05-24 | 2019-08-23 | 北京立群建筑科学研究院 | Environment detection method, device, equipment and storage medium |
CN111860997A (en) * | 2020-07-17 | 2020-10-30 | 海南大学 | Cross-data, information, knowledge modality and dimension early warning method and component |
CN111860997B (en) * | 2020-07-17 | 2021-12-28 | 海南大学 | Cross-data, information, knowledge modality and dimension early warning method and component |
CN113011098A (en) * | 2021-03-26 | 2021-06-22 | 云南电网有限责任公司电力科学研究院 | Power transmission corridor area lightning activity rule analysis display model and system |
CN113011098B (en) * | 2021-03-26 | 2023-09-01 | 云南电网有限责任公司电力科学研究院 | Lightning activity rule analysis display model and system for power transmission corridor area |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108764550A (en) | Lightning Warning method and system based on transmission line information data | |
Chiang et al. | Comparison of static-feedforward and dynamic-feedback neural networks for rainfall–runoff modeling | |
CN102629294B (en) | Probability evaluation method of failure caused by typhoon to power transmission line | |
CN108154651A (en) | A kind of device and method for landslide disaster Monitoring and forecasting system in real-time | |
CN109272146A (en) | A kind of Forecasting Flood method corrected based on deep learning model and BP neural network | |
CN110263866A (en) | A kind of power consumer load setting prediction technique based on deep learning | |
CN104992373B (en) | A kind of transmission line of electricity natural hybridized orbit method for early warning | |
Minglei et al. | Classified real-time flood forecasting by coupling fuzzy clustering and neural network | |
CN107169645B (en) | Power transmission line fault probability online evaluation method considering influence of rainstorm disaster | |
CN105117602A (en) | Metering apparatus operation state early warning method | |
CN106017551A (en) | Intelligent transmission line integrated monitoring analysis and early warning method | |
CN106682774A (en) | Contact net insulator pollution flashover prediction method | |
CN104318347A (en) | Power transmission line icing state assessment method based on information fusion of multiple sensors | |
CN109118020A (en) | A kind of subway station energy consumption short term prediction method and its forecasting system | |
CN104732058A (en) | Method for assessing state of multi-dimensional power transmission equipment | |
CN107069975A (en) | A kind of distributed transmission facility status data reponse system and its method | |
Hsieh et al. | A hybrid approach of artificial neural network and multiple regression to forecast typhoon rainfall and groundwater-level change | |
CN116090821A (en) | Power distribution network line security risk assessment method considering extreme weather | |
CN108694479A (en) | Consider the distribution network reliability prediction technique that weather influences time between overhaul | |
CN104050518B (en) | Power grid convection disaster-causing strong wind early warning method based on Doppler weather radar | |
Kumar | Market clearing price prediction using ANN in Indian electricity markets | |
CN105427189A (en) | Power grid vulnerability evaluation method under landslide hazard | |
CN110533243A (en) | A kind of building energy consumption prediction technique based on fractal theory | |
CN112434887B (en) | Water supply network risk prediction method combining network kernel density estimation and SVM | |
CN112257329A (en) | Method for judging influence of typhoon on line |
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: 20181106 |
|
RJ01 | Rejection of invention patent application after publication |