CN112949515A - Line forest fire early warning method and system based on monitoring information - Google Patents
Line forest fire early warning method and system based on monitoring information Download PDFInfo
- Publication number
- CN112949515A CN112949515A CN202110255791.2A CN202110255791A CN112949515A CN 112949515 A CN112949515 A CN 112949515A CN 202110255791 A CN202110255791 A CN 202110255791A CN 112949515 A CN112949515 A CN 112949515A
- Authority
- CN
- China
- Prior art keywords
- probability
- month
- mountain fire
- hour
- order
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 154
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000009826 distribution Methods 0.000 claims description 73
- 238000012545 processing Methods 0.000 claims description 7
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 description 7
- 239000000779 smoke Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/10—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/12—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
- G08B17/125—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions by using a video camera to detect fire or smoke
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- Mathematical Analysis (AREA)
- Emergency Management (AREA)
- Pure & Applied Mathematics (AREA)
- Business, Economics & Management (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Operations Research (AREA)
- Algebra (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a line forest fire early warning method and system based on monitoring information, wherein monitoring time and a monitoring area identifier are input into a target probability prediction model to obtain a prediction probability, and when the prediction probability reaches an early warning value and relevant characteristics exist among monitoring information characteristics of lines of various voltage classes, the prediction probability and the monitoring information characteristics are subjected to logical operation to obtain a first logical operation result; and finally, if one of the first logic operation result and the second logic operation result is true, performing line mountain fire alarm to realize remote line mountain fire early warning of ground-air linkage and protect the safe, stable and economic operation of a power grid line.
Description
Technical Field
The invention relates to the technical field of power grid line safety monitoring, in particular to a line forest fire early warning method and system based on monitoring information.
Background
Because most of the power transmission lines are erected in the field, the operation environment is complex and changeable, and faults are easy to occur due to the influence of disastrous weather such as thunder and lightning, mountain fire, ice disasters and the like. Especially, with the continuous development of power grids, power transmission lines often span mountain areas with luxuriant vegetation, large-scale mountain fire disasters nearby line corridors often occur, when the mountain fire disasters are outbreaked in a large range, threats are often generated on multiple lines, and for tidal current sections with heavier loads, cascading tripping accidents can be caused after the lines are tripped due to mountain fire, so that the power grids are powered off in a large scale. Or, the power transmission line is easy to swing when meeting wind and frequently contacts trees around the power transmission line or the slope of a mountain, so that short circuit discharge of the power transmission line is caused to cause a forest fire.
In view of the fact that mountain fire has serious threats to the safe and stable operation of a power grid for several times in recent years, the following methods are adopted at present: the personnel stationing patrol consumes manpower, material resources and financial resources, and the patrol range is limited; the satellite image identification-based means is limited to meteorological interference, low resolution, satellite earth-surrounding period and other reasons, so that the fire cannot be found in time, particularly, when the fire area is too small, early warning cannot be carried out at the first time, and the optimal time for fire extinguishment is missed for the fire caused by discharge of short circuit, disconnection, grounding and the like of a power transmission line; the additional installation of smoke sensing systems, infrared cameras and the like greatly increases the expense, and the operation and maintenance of the systems in remote areas are difficult to realize.
Disclosure of Invention
The invention aims to solve the technical problems that the existing manual inspection, satellite image recognition and smoke sensor added for mountain fire safety monitoring of a power transmission line supposed to be in a mountain forest consume manpower, fund and interference are large and timely recognition cannot be realized, so that the invention provides a line mountain fire early warning method based on monitoring information.
The invention is realized by the following technical scheme:
a line forest fire early warning method based on monitoring information comprises the following steps:
acquiring monitoring time and a monitoring area identifier, and inputting the monitoring time and the monitoring area identifier into a target probability prediction model to obtain a prediction probability;
acquiring the monitoring information characteristics of each voltage grade circuit in the corresponding area in real time according to the monitoring area identification;
when the prediction probability reaches an early warning value and relevant features exist among the monitoring information features of the circuits of all voltage classes, carrying out logic operation on the prediction probability and the monitoring information features to obtain a first logic operation result;
acquiring satellite image recognition early warning information corresponding to the monitoring area identification, and performing logical operation on the monitoring information characteristics and the satellite image recognition early warning information to obtain a second logical operation result;
and if one of the first logical operation result and the second logical operation result is true, performing line forest fire alarm.
Further, establishing the target probability prediction model includes:
counting historical mountain fire data corresponding to different monitoring area identifications by taking the year as a unit;
counting the distribution times of the historical mountain fire data corresponding to each monitoring area identifier in each month of the year and the distribution times of the historical mountain fire data in each hour of the day based on the historical mountain fire data;
calculating the mountain fire month distribution probability corresponding to each monitoring area identifier through a month distribution probability function to obtain month probability distribution data, performing curve fitting on the month probability distribution data by adopting a fitting curve function order successive increasing method to obtain a month fitting curve function, determining the order of the month fitting curve function, and establishing a month probability prediction model;
calculating the hill fire hour distribution probability corresponding to each monitoring area identification through an hour distribution probability function to obtain hour probability distribution data, performing curve fitting on the hour probability distribution data by adopting a fitting curve function order successive increasing method to obtain an hour fitting curve function, determining the order of the hour fitting curve function, and establishing an hour probability prediction model;
and multiplying the month probability prediction model and the hour probability prediction model to obtain a target probability prediction model.
Further, determining an order of the month fit curve function comprises:
initializing the order of the month fitting curve function to obtain a month initial order;
sequentially inputting the initial order of the months and the number of the months into a month fitting curve function according to the sequence from small to large, and calculating the mountain fire probability fitting value of each month;
obtaining a mountain fire historical probability value corresponding to each month, and calculating the mountain fire probability fitting value and the mountain fire historical probability value of each month through a loss function to obtain a mountain fire probability loss value of each month;
and when the probability loss values of the forest fires of all the months are not smaller than the preset precision threshold, adding one to the initial order of the month, and stopping taking the corresponding order as the order of the month fitting curve function until the probability loss values of the forest fires of all the months are smaller than the preset precision threshold.
Further, determining the order of the hour-fit curve function comprises:
initializing the order of the hour fitting curve function to obtain an hour initial order;
sequentially inputting the hour initial order and the hours into an hour fitting curve function according to the order from hour to hour, and calculating the mountain fire probability fitting value of each hour;
acquiring a mountain fire historical probability value corresponding to each hour, and calculating a mountain fire probability fitting value and a mountain fire historical probability value of each hour through a loss function to obtain a mountain fire probability loss value of each hour;
and when the mountain fire probability loss values of all the hours are not smaller than a preset precision threshold, adding one to the initial order of the hour, and stopping taking the corresponding order as the order of the hour fitting curve function until the mountain fire probability loss values of all the hours are smaller than the preset precision threshold.
Further, the loss function is specifically:
wherein, YiThe probability value of the mountain fire history is shown,
representing the probability of a mountain fire, xiAnd (3) representing the distribution times of the forest fire at the ith time point, wherein i comprises months and hours, n represents the order, and a represents a coefficient obtained by calculation according to a fitted curve function.
Further, acquiring monitoring information characteristics of each voltage class line in the corresponding area includes:
acquiring the characteristic signal names of lines with different voltage levels corresponding to monitoring time in each monitoring area and counting the characteristic signal quantity of each characteristic signal name;
and when the number of the characteristic signals corresponding to the characteristic signal name at the monitoring time is constantly larger than the number of the characteristic signals corresponding to the characteristic signal name in the adjacent time period, taking the characteristic signal name as the monitoring information characteristic.
Further, the line forest fire early warning method based on the monitoring information further comprises the following steps:
and automatically acquiring monitoring time and monitoring area identification according to the preset interval time.
A line forest fire early warning system based on monitoring information comprises:
the target probability prediction model identification module is used for acquiring monitoring time and monitoring area identification, and inputting the monitoring time and the monitoring area identification into the target probability prediction model to obtain prediction probability;
the monitoring information characteristic acquisition module is used for acquiring the monitoring information characteristics of each voltage grade circuit in the corresponding area in real time according to the monitoring area identification;
the first logic operation result acquisition module is used for carrying out logic operation on the prediction probability and the monitoring information characteristics when the prediction probability reaches an early warning value and the monitoring information characteristics of each voltage level circuit have related characteristics, so as to acquire a first logic operation result;
the second logic operation result acquisition module is used for acquiring satellite image identification early warning information corresponding to the monitoring area identification, and performing logic operation on the monitoring information characteristics and the satellite image identification early warning information to obtain a second logic operation result;
and the line forest fire alarm module is used for alarming line forest fire if one of the first logical operation result and the second logical operation result is true.
Further, the target probability prediction model establishing unit includes:
the historical mountain fire data acquisition unit is used for counting historical mountain fire data corresponding to different monitoring area identifications in units of years;
the historical mountain fire data processing unit is used for counting the distribution times of the historical mountain fire data corresponding to each monitoring area identifier in each month in one year and the distribution times of the historical mountain fire data in each hour in one day based on the historical mountain fire data;
the month probability prediction model establishing unit is used for calculating the mountain fire month distribution probability corresponding to each monitoring area identifier through a month distribution probability function to obtain month probability distribution data, performing curve fitting on the month probability distribution data by adopting a fitting curve function order successive increasing method to obtain a month fitting curve function, determining the order of the month fitting curve function, and establishing a month probability prediction model;
and the hour probability prediction model establishing unit is used for calculating the mountain fire hour distribution probability corresponding to each monitoring area identifier through an hour distribution probability function to obtain hour probability distribution data, performing curve fitting on the hour probability distribution data by adopting a fitting curve function order successive increasing method to obtain an hour fitting curve function, determining the order of the hour fitting curve function, and establishing the hour probability prediction model.
And the target probability prediction model processing unit is used for multiplying the month probability prediction model and the hour probability prediction model to obtain a target probability prediction model.
Further, the month probability prediction model establishing unit includes:
the month order initialization unit is used for initializing the order of the month fitting curve function to obtain a month initial order;
the month and mountain fire probability fitting value calculating unit is used for sequentially inputting the initial order of the months and the number of the months into a month fitting curve function according to the sequence from small to large and calculating the mountain fire probability fitting value of each month;
the moon mountain fire probability loss value calculation unit is used for acquiring mountain fire historical probability values corresponding to the months, and calculating the mountain fire probability fitting values and the mountain fire historical probability values of the months through a loss function to obtain mountain fire probability loss values of the months;
and the month fitting curve order determining unit is used for adding one to the initial month order when the mountain fire probability loss values of all the months are not smaller than the preset precision threshold, and stopping taking the corresponding order as the order of the month fitting curve function until the mountain fire probability loss values of all the months are smaller than the preset precision threshold.
The invention provides a line forest fire early warning method and system based on monitoring information, wherein monitoring time and a monitoring area identifier are input into a target probability prediction model to obtain a prediction probability, and when the prediction probability reaches an early warning value and relevant characteristics exist among monitoring information characteristics of lines of various voltage levels, logic operation is carried out on the prediction probability and the monitoring information characteristics to obtain a first logic operation result; and finally, if one of the first logic operation result and the second logic operation result is true, performing line mountain fire alarm to realize remote line mountain fire early warning of ground-air linkage and protect the safe, stable and economic operation of a power grid line.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principles of the invention. In the drawings:
fig. 1 is a flowchart of a line forest fire early warning method based on monitoring information according to the present invention.
Fig. 2 is a specific flowchart of step S10 in fig. 1.
Fig. 3 is a specific flowchart of step S103 in fig. 2.
Fig. 4 is a specific flowchart of step S104 in fig. 2.
Fig. 5 is a specific flowchart of step S20 in fig. 1.
Fig. 6 is a schematic diagram of a line forest fire early warning system based on monitoring information according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to examples and accompanying drawings, and the exemplary embodiments and descriptions thereof are only used for explaining the present invention and are not meant to limit the present invention.
Example 1
As shown in fig. 1, the present invention provides a line forest fire early warning method based on monitoring information, which includes:
s10: and acquiring monitoring time and a monitoring area identifier, and inputting the monitoring time and the monitoring area identifier into a target probability prediction model to obtain a prediction probability.
Specifically, in this embodiment, the manner of acquiring the monitoring time and the monitoring area identifier may be manually input, or may be automatically acquired according to a preset interval time.
S20: and acquiring the monitoring information characteristics of each voltage grade circuit in the corresponding area in real time according to the monitoring area identification.
Specifically, the voltage class lines in the present embodiment include different voltage class lines (e.g., 500kV, 220kV, and 110kV) in the main network and different voltage class lines (e.g., 35kV, 10kV, and 6kV) in the distribution network.
S30: and when the prediction probability reaches the early warning value and relevant characteristics exist among the monitoring information characteristics of the circuits of all voltage classes, carrying out logic operation on the prediction probability and the monitoring information characteristics to obtain a first logic operation result.
Specifically, since the device configuration characteristics of the lines with different voltage classes and different areas are different, a signal with an obvious characteristic needs to be selected as the monitoring information characteristic. The characteristic signals include, but are not limited to, fault recording starting signals, voltage out-of-limit signals, harmonic elimination device action signals, traveling wave distance measurement starting, grounding characteristic signals, distance protection starting signals, device abnormity, zero sequence related characteristic signals and accident brake separating signals.
The first logic operation result refers to a result obtained by performing logic and operation on the prediction probability and the monitoring information characteristic.
S40: and acquiring satellite image recognition early warning information corresponding to the monitoring area identifier, and performing logical operation on the monitoring information characteristics and the satellite image recognition early warning information to obtain a second logical operation result.
And the second logical operation result refers to a result obtained by logical and operation of the monitoring information characteristic and the satellite image recognition early warning information.
Specifically, the satellite image identification data may be acquired once at a preset time interval (e.g., 10 minutes), or may be acquired at a time defined by a user.
S50: and if one of the first logical operation result and the second logical operation result is true, performing line forest fire alarm.
Specifically, the early warning value P of the probability model based on the historical data space-time distributionAlarmDefault settings are 100:
P≥PAlarm→bool:1
P＜PAlarm→bool:0
when the probability value P based on the historical data space-time distribution is larger than the early warning value PAlarmIf so, outputting Boolean quantity to be 1; otherwise, the output boolean quantity is 0.
When acquiring monitoring information characteristics of each voltage grade line
Cyclically monitoring the characteristic signal Ti→bool:1
Not monitoring the characteristic signal Ti→bool:0
When the satellite image identifies the early warning information
Recognition early warning S → bol: 1 for monitoring satellite image
No monitoring of satellite image recognition early warning S → bol: 0
The final remote early warning output logic of ground-air linked line and mountain fire is as follows:
when P is present&&Ti1 → pool: 1, output the early warning
When P is present&&Ti0 → bol: 0, no warning output is made as S&&Ti1 → pool: 1, output the early warning
When S is&&TiNo warning output is made 0 → bol 0
When early warning output is carried out, the line forest fire alarm is carried out; and when the early warning output is not carried out, the line mountain fire alarm is not carried out. The prediction probability is combined with the monitoring information characteristic, so that the interference of the monitoring information characteristic quantity in a low-risk time period and a low-risk area can be avoided; the monitoring information characteristics are combined with the satellite image recognition early warning information, so that special fires occurring in low-risk time periods and low-risk areas can be avoided, and the remote early warning of ground-air linked line mountain fires is finally realized.
Further, as shown in fig. 2, the establishing of the target probability prediction model in this embodiment specifically includes the following steps:
s101: and counting historical mountain fire data corresponding to different monitoring area identifications by taking the year as a unit.
S102: and counting the distribution times of the historical mountain fire data corresponding to each monitoring area identification in each month of the year and in each hour of the day based on the historical mountain fire data.
Specifically, the historical mountain fire data of the power grid line of the monitoring area identifier a is counted by taking the year as a unit, and as shown in table 1, the historical mountain fire data includes mountain fire distribution times of each month in the past year corresponding to the monitoring area identifier a.
TABLE 1
Wherein, the distribution frequency Xm of mountain fire in the region A at the i-th month is shown in Table 1, wherein (i is not less than 1 and not more than 12, i belongs to N).
And counting the historical mountain fire data of the power grid line of the monitoring area identifier A by taking the hour as a unit, wherein the historical mountain fire data comprises the mountain fire distribution times of each hour corresponding to the monitoring area identifier A shown in the table 2.
TABLE 2
Wherein, the distribution frequency Xh of mountain fire in the region A at the ith hour is shown in Table 2, wherein (i is more than or equal to 1 and less than or equal to 12, and i belongs to N).
S103: calculating the mountain fire month distribution probability corresponding to each monitoring area identifier through a month distribution probability function to obtain month probability distribution data, performing curve fitting on the month probability distribution data by adopting a fitting curve function order successive increasing method to obtain a month fitting curve function, determining the order of the month fitting curve function, and establishing a month probability prediction model.
S104: calculating the hill fire hour distribution probability corresponding to each monitoring area identification through an hour distribution probability function to obtain hour probability distribution data, performing curve fitting on the hour probability distribution data by adopting a fitting curve function order successive increasing method to obtain an hour fitting curve function, determining the order of the hour fitting curve function, and establishing an hour probability prediction model.
Specifically, the month distribution probability function is:
wherein λ is1Is the relaxation factor of the month.
The hour distribution probability function is:
wherein λ is2Is the relaxation factor of the month.
In order to avoid that the result obtained by multiplying the probability values (between 0 and 1) calculated by the hour distribution probability function is too small to be convenient for judgment and observation, the relaxation coefficient is set to be 100 by default.
The fitted curve function is expressed by polyfit (x, y, n), and specifically: f (x, y, n) is polyfit (x, y, n), i.e. y is anxn+an-1xn-1+…+a0Wherein n represents an order, x represents a month or an hour, y represents a probability value corresponding to each month or each hour, and a represents a coefficient calculated according to a fitted curve function.
S105: and multiplying the month probability prediction model and the hour probability prediction model to obtain a target probability prediction model.
Specifically, the total probability P (a value between 0 and 10000), that is, the regional forest fire prediction probability based on the temporal-spatial distribution of the historical data, is obtained by multiplying the probability value calculated by the month probability prediction model by the hour probability prediction model probability value. Setting the pre-warning value to PAlarm(default is set to be 100, and adjustment can be carried out according to actual conditions), and when the early warning value is reached, the Boolean quantity of the early warning value is set to be 1.
The target probability prediction model obtained by multiplying the month probability prediction model and the hour probability prediction model is a probability model continuously calculated according to real-time.
Further, as shown in fig. 3, in step S103, determining an order of the month fitting curve function specifically includes the following steps:
s1031: and initializing the order of the month fitting curve function to obtain the initial order of the month.
S1032: and sequentially inputting the initial order of the months and the number of the months into a month fitting curve function according to the sequence from small to large, and calculating the mountain fire probability fitting value of each month.
S1033: and obtaining the mountain fire historical probability value corresponding to each month, and calculating the mountain fire probability fitting value and the mountain fire historical probability value of each month through a loss function to obtain the mountain fire probability loss value of each month.
S1034: and when the probability loss values of the forest fires of all the months are not smaller than the preset precision threshold, adding one to the initial order of the month until the probability loss values of the forest fires of all the months are smaller than the preset precision threshold, and stopping taking the corresponding order as the order of the month fitting curve function.
Further, the loss function in this embodiment is specifically:
wherein, YiThe probability value of the mountain fire history is shown,
representing the probability of a mountain fire, xiIndicating the number of mountain fire distributions at the ith time point, i including months and hours,nthe order of the order is represented,arepresenting the coefficients calculated from the fitted curve function.
In order to minimize the loss function L, the partial derivatives are calculated for the coefficients of the loss function such that the value of the partial derivative is 0, which can be calculated by the following formula.
Finally, the coefficients that minimize the loss function L, i.e., the coefficients that fit the curve function polyfit (x, y, n), are obtained by calculation.
Further, as shown in fig. 4, in step S104, determining the order of the hour-fitting curve function specifically includes the following steps:
s1041: and initializing the order of the hour fitting curve function to obtain an hour initial order.
S1042: and sequentially inputting the initial order of the hours and the hours into an hour fitting curve function according to the order from small to large, and calculating the mountain fire probability fitting value of each hour.
S1043: and acquiring a mountain fire historical probability value corresponding to each hour, and calculating the mountain fire probability fitting value and the mountain fire historical probability value of each hour through a loss function to obtain a mountain fire probability loss value of each hour.
S1044: and when the mountain fire probability loss values of all the hours are not smaller than the preset precision threshold, adding one to the initial order of the hours, and stopping taking the corresponding order as the order of the hour fitting curve function until the mountain fire probability loss values of all the hours are smaller than the preset precision threshold.
Further, as shown in fig. 5, in step S20, acquiring the monitoring information characteristics of each voltage class line in the corresponding area specifically includes the following steps:
s21: and acquiring the characteristic signal names of the lines with different voltage levels corresponding to the monitoring time in each monitoring area and counting the characteristic signal quantity of each characteristic signal name.
S22: and when the number of the characteristic signals corresponding to the characteristic signal name at the monitoring time is constantly larger than the number of the characteristic signals corresponding to the characteristic signal name in the adjacent time period, taking the characteristic signal name as the monitoring information characteristic.
Specifically, by way of disclosure, may be represented as:
wherein i represents the number of the characteristic signal, j represents the number of days of the statistical time periodThe number m is required to be odd, and the number of the day when the mountain fire occurs is the median value of the number, namely
Example 2
As shown in fig. 6, this embodiment provides a monitoring information-based line forest fire early warning system corresponding to the monitoring information-based line forest fire early warning method of embodiment 1 one to one, including:
and the target probability prediction model identification module 10 is used for acquiring the monitoring time and the monitoring area identifier, and inputting the monitoring time and the monitoring area identifier into the target probability prediction model to obtain the prediction probability.
And the monitoring information characteristic obtaining module 20 is configured to obtain, in real time, monitoring information characteristics of each voltage class line in the corresponding area according to the monitoring area identifier.
And a first logic operation result obtaining module 30, configured to perform logic operation on the prediction probability and the monitoring information features to obtain a first logic operation result when the prediction probability reaches the early warning value and there is a relevant feature between the monitoring information features of the voltage class lines.
And the second logical operation result acquisition module 40 is configured to acquire satellite image recognition early warning information corresponding to the monitored area identifier, and perform logical operation on the monitoring information characteristics and the satellite image recognition early warning information to obtain a second logical operation result.
And the line forest fire alarm module 50 is used for alarming line forest fire if one of the first logical operation result and the second logical operation result is true.
Further, the target probability prediction model establishing unit comprises a historical forest fire data acquiring unit, a historical forest fire data processing unit, a month probability prediction model establishing unit and an hour probability prediction model establishing unit.
And the historical mountain fire data acquisition unit is used for counting the historical mountain fire data corresponding to different monitoring area identifiers by taking the year as a unit.
And the historical mountain fire data processing unit is used for counting the distribution times of the historical mountain fire data corresponding to each monitoring area identifier in each month in the year and the distribution times in each hour in the day based on the historical mountain fire data.
And the month probability prediction model establishing unit is used for calculating the mountain fire month distribution probability corresponding to each monitoring area identifier through the month distribution probability function to obtain month probability distribution data, performing curve fitting on the month probability distribution data by adopting a fitting curve function order successive increasing method to obtain a month fitting curve function, determining the order of the month fitting curve function, and establishing the month probability prediction model.
And the hour probability prediction model establishing unit is used for calculating the mountain fire hour distribution probability corresponding to each monitoring area identifier through an hour distribution probability function to obtain hour probability distribution data, performing curve fitting on the hour probability distribution data by adopting a fitting curve function order successive increasing method to obtain an hour fitting curve function, determining the order of the hour fitting curve function, and establishing the hour probability prediction model.
And the target probability prediction model processing unit is used for multiplying the month probability prediction model and the hour probability prediction model to obtain a target probability prediction model.
Further, the month probability prediction model establishing unit comprises a month order initializing unit, a month and mountain fire probability fitting value calculating unit, a month and mountain fire probability loss value calculating unit and a month fitting curve order determining unit.
And the month order initialization unit is used for initializing the order of the month fitting curve function to obtain the month initial order.
And the moon and mountain fire probability fitting value calculating unit is used for sequentially inputting the initial order of the months and the number of the months into the month fitting curve function according to the sequence from small to large and calculating the mountain fire probability fitting value of each month.
And the moon mountain fire probability loss value calculating unit is used for acquiring the mountain fire historical probability value corresponding to each month, and calculating the mountain fire probability fitting value and the mountain fire historical probability value of each month through a loss function to obtain the mountain fire probability loss value of each month.
And the month fitting curve order determining unit is used for adding one to the initial month order when the mountain fire probability loss values of all the months are not smaller than the preset precision threshold, stopping the operation until the mountain fire probability loss values of all the months are smaller than the preset precision threshold, and taking the corresponding order as the order of the month fitting curve function.
The above embodiments are provided to further explain the objects, technical solutions and advantages of the present invention in detail, it should be understood that the above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the scope of the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A line forest fire early warning method based on monitoring information is characterized by comprising the following steps:
acquiring monitoring time and a monitoring area identifier, and inputting the monitoring time and the monitoring area identifier into a target probability prediction model to obtain a prediction probability;
acquiring the monitoring information characteristics of each voltage grade circuit in the corresponding area in real time according to the monitoring area identification;
when the prediction probability reaches an early warning value and relevant features exist among the monitoring information features of the circuits of all voltage classes, carrying out logic operation on the prediction probability and the monitoring information features to obtain a first logic operation result;
acquiring satellite image recognition early warning information corresponding to the monitoring area identification, and performing logical operation on the monitoring information characteristics and the satellite image recognition early warning information to obtain a second logical operation result;
and if one of the first logical operation result and the second logical operation result is true, performing line forest fire alarm.
2. The method for early warning of the road and mountain fire based on the monitoring information as claimed in claim 1, wherein the establishing of the target probability prediction model comprises:
counting historical mountain fire data corresponding to different monitoring area identifications by taking the year as a unit;
counting the distribution times of the historical mountain fire data corresponding to each monitoring area identifier in each month of the year and the distribution times of the historical mountain fire data in each hour of the day based on the historical mountain fire data;
calculating the mountain fire month distribution probability corresponding to each monitoring area identifier through a month distribution probability function to obtain month probability distribution data, performing curve fitting on the month probability distribution data by adopting a fitting curve function order successive increasing method to obtain a month fitting curve function, determining the order of the month fitting curve function, and establishing a month probability prediction model;
calculating the hill fire hour distribution probability corresponding to each monitoring area identification through an hour distribution probability function to obtain hour probability distribution data, performing curve fitting on the hour probability distribution data by adopting a fitting curve function order successive increasing method to obtain an hour fitting curve function, determining the order of the hour fitting curve function, and establishing an hour probability prediction model;
and multiplying the month probability prediction model and the hour probability prediction model to obtain a target probability prediction model.
3. The method of claim 2, wherein determining the order of the month fitting curve function comprises:
initializing the order of the month fitting curve function to obtain a month initial order;
sequentially inputting the initial order of the months and the number of the months into a month fitting curve function according to the sequence from small to large, and calculating the mountain fire probability fitting value of each month;
obtaining a mountain fire historical probability value corresponding to each month, and calculating the mountain fire probability fitting value and the mountain fire historical probability value of each month through a loss function to obtain a mountain fire probability loss value of each month;
and when the probability loss values of the forest fires of all the months are not smaller than the preset precision threshold, adding one to the initial order of the month, and stopping taking the corresponding order as the order of the month fitting curve function until the probability loss values of the forest fires of all the months are smaller than the preset precision threshold.
4. The line forest fire early warning method based on monitoring information as claimed in claim 2, wherein the determining the order of the hour fitting curve function comprises:
initializing the order of the hour fitting curve function to obtain an hour initial order;
sequentially inputting the hour initial order and the hours into an hour fitting curve function according to the order from hour to hour, and calculating the mountain fire probability fitting value of each hour;
acquiring a mountain fire historical probability value corresponding to each hour, and calculating a mountain fire probability fitting value and a mountain fire historical probability value of each hour through a loss function to obtain a mountain fire probability loss value of each hour;
and when the mountain fire probability loss values of all the hours are not smaller than a preset precision threshold, adding one to the initial order of the hour, and stopping taking the corresponding order as the order of the hour fitting curve function until the mountain fire probability loss values of all the hours are smaller than the preset precision threshold.
5. The line and mountain fire early warning method based on monitoring information according to claim 3 or 4, wherein the loss function is specifically:
wherein, YiThe probability value of the mountain fire history is shown,
representing the probability of a mountain fire, xiShowing the number of times of distribution of mountain fire at the ith time point, i includesMonth and hour, n represents the order, and a represents the coefficient calculated from the fitted curve function.
6. The line forest fire early warning method based on monitoring information according to claim 1, wherein the obtaining of the monitoring information characteristics of each voltage level line in the corresponding area comprises:
acquiring the characteristic signal names of lines with different voltage levels corresponding to monitoring time in each monitoring area and counting the characteristic signal quantity of each characteristic signal name;
and when the number of the characteristic signals corresponding to the characteristic signal name at the monitoring time is constantly larger than the number of the characteristic signals corresponding to the characteristic signal name in the adjacent time period, taking the characteristic signal name as the monitoring information characteristic.
7. The line forest fire early warning method based on the monitoring information as claimed in claim 1, wherein the line forest fire early warning method based on the monitoring information further comprises:
and automatically acquiring monitoring time and monitoring area identification according to the preset interval time.
8. The utility model provides a circuit mountain fire early warning system based on monitoring information which characterized in that includes:
the target probability prediction model identification module is used for acquiring monitoring time and monitoring area identification, and inputting the monitoring time and the monitoring area identification into the target probability prediction model to obtain prediction probability;
the monitoring information characteristic acquisition module is used for acquiring the monitoring information characteristics of each voltage grade circuit in the corresponding area in real time according to the monitoring area identification;
the first logic operation result acquisition module is used for carrying out logic operation on the prediction probability and the monitoring information characteristics when the prediction probability reaches an early warning value and the monitoring information characteristics of each voltage level circuit have related characteristics, so as to acquire a first logic operation result;
the second logic operation result acquisition module is used for acquiring satellite image identification early warning information corresponding to the monitoring area identification, and performing logic operation on the monitoring information characteristics and the satellite image identification early warning information to obtain a second logic operation result;
and the line forest fire alarm module is used for alarming line forest fire if one of the first logical operation result and the second logical operation result is true, and if the two logical operation results are true.
9. The system of claim 8, wherein the target probability prediction model building unit comprises:
the historical mountain fire data acquisition unit is used for counting historical mountain fire data corresponding to different monitoring area identifications in units of years;
the historical mountain fire data processing unit is used for counting the distribution times of the historical mountain fire data corresponding to each monitoring area identifier in each month in one year and the distribution times of the historical mountain fire data in each hour in one day based on the historical mountain fire data;
the month probability prediction model establishing unit is used for calculating the mountain fire month distribution probability corresponding to each monitoring area identifier through a month distribution probability function to obtain month probability distribution data, performing curve fitting on the month probability distribution data by adopting a fitting curve function order successive increasing method to obtain a month fitting curve function, determining the order of the month fitting curve function, and establishing a month probability prediction model;
the hourly probability prediction model establishing unit is used for calculating the mountain fire hourly distribution probability corresponding to each monitoring area identifier through an hourly distribution probability function to obtain hourly probability distribution data, performing curve fitting on the hourly probability distribution data by adopting a fitting curve function order successive increasing method to obtain an hourly fitting curve function, determining the order of the hourly fitting curve function, and establishing an hourly probability prediction model;
and the target probability prediction model processing unit is used for multiplying the month probability prediction model and the hour probability prediction model to obtain a target probability prediction model.
10. The system of claim 9, wherein the month probability prediction model building unit comprises:
the month order initialization unit is used for initializing the order of the month fitting curve function to obtain a month initial order;
the month and mountain fire probability fitting value calculating unit is used for sequentially inputting the initial order of the months and the number of the months into a month fitting curve function according to the sequence from small to large and calculating the mountain fire probability fitting value of each month;
the moon mountain fire probability loss value calculation unit is used for acquiring mountain fire historical probability values corresponding to the months, and calculating the mountain fire probability fitting values and the mountain fire historical probability values of the months through a loss function to obtain mountain fire probability loss values of the months;
and the month fitting curve order determining unit is used for adding one to the initial month order when the mountain fire probability loss values of all the months are not smaller than the preset precision threshold, and stopping taking the corresponding order as the order of the month fitting curve function until the mountain fire probability loss values of all the months are smaller than the preset precision threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110255791.2A CN112949515B (en) | 2021-03-09 | 2021-03-09 | Line forest fire early warning method and system based on monitoring information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110255791.2A CN112949515B (en) | 2021-03-09 | 2021-03-09 | Line forest fire early warning method and system based on monitoring information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112949515A true CN112949515A (en) | 2021-06-11 |
| CN112949515B CN112949515B (en) | 2022-07-08 |
Family
ID=76228553
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110255791.2A Active CN112949515B (en) | 2021-03-09 | 2021-03-09 | Line forest fire early warning method and system based on monitoring information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112949515B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973584A (en) * | 2022-05-10 | 2022-08-30 | 云南电网有限责任公司电力科学研究院 | Mountain fire warning method and device, computer equipment and storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040061777A1 (en) * | 2002-05-20 | 2004-04-01 | Mokhtar Sadok | Detecting fire using cameras |
| TW201028948A (en) * | 2009-01-16 | 2010-08-01 | Lin jin sui | Fire risk quantification system and method |
| KR20100118368A (en) * | 2009-04-28 | 2010-11-05 | 부산대학교 산학협력단 | Fire detecting method using hidden markov models in video surveillance and monitoring system |
| CN103886385A (en) * | 2014-02-20 | 2014-06-25 | 中国林业科学研究院森林生态环境与保护研究所 | Method for predicting forest fire hazard day occurrence probability |
| CN106022622A (en) * | 2016-05-25 | 2016-10-12 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Mountain fire risk assessment method of transmission line |
| CN107067101A (en) * | 2017-02-16 | 2017-08-18 | 湖南省湘电试研技术有限公司 | Many fire point power grid risks minimize emergence treating method and system |
| CN107067683A (en) * | 2017-04-14 | 2017-08-18 | 湖南省湘电试研技术有限公司 | A kind of transmission line forest fire clusters quantitative forecast method and system |
| CN108416968A (en) * | 2018-01-31 | 2018-08-17 | 国家能源投资集团有限责任公司 | Fire alarm method and apparatus |
| CN109119976A (en) * | 2018-09-05 | 2019-01-01 | 国网湖南省电力有限公司 | The automatic adjusting method and system of relay protection constant value for transmission line forest fire |
| CN109118001A (en) * | 2018-08-09 | 2019-01-01 | 成都天地量子科技有限公司 | A kind of mountain fire monitoring method and system based on satellite remote sensing date |
| CN109377701A (en) * | 2018-10-29 | 2019-02-22 | 国网四川省电力公司电力科学研究院 | The real-time early warning system and method for transmission line forest fire fire behavior development trend |
| WO2020083091A1 (en) * | 2018-10-25 | 2020-04-30 | 国网湖南省电力有限公司 | Risk analysis method and system for power grid fault |
-
2021
- 2021-03-09 CN CN202110255791.2A patent/CN112949515B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040061777A1 (en) * | 2002-05-20 | 2004-04-01 | Mokhtar Sadok | Detecting fire using cameras |
| TW201028948A (en) * | 2009-01-16 | 2010-08-01 | Lin jin sui | Fire risk quantification system and method |
| KR20100118368A (en) * | 2009-04-28 | 2010-11-05 | 부산대학교 산학협력단 | Fire detecting method using hidden markov models in video surveillance and monitoring system |
| CN103886385A (en) * | 2014-02-20 | 2014-06-25 | 中国林业科学研究院森林生态环境与保护研究所 | Method for predicting forest fire hazard day occurrence probability |
| CN106022622A (en) * | 2016-05-25 | 2016-10-12 | 中国南方电网有限责任公司超高压输电公司检修试验中心 | Mountain fire risk assessment method of transmission line |
| CN107067101A (en) * | 2017-02-16 | 2017-08-18 | 湖南省湘电试研技术有限公司 | Many fire point power grid risks minimize emergence treating method and system |
| CN107067683A (en) * | 2017-04-14 | 2017-08-18 | 湖南省湘电试研技术有限公司 | A kind of transmission line forest fire clusters quantitative forecast method and system |
| CN108416968A (en) * | 2018-01-31 | 2018-08-17 | 国家能源投资集团有限责任公司 | Fire alarm method and apparatus |
| CN109118001A (en) * | 2018-08-09 | 2019-01-01 | 成都天地量子科技有限公司 | A kind of mountain fire monitoring method and system based on satellite remote sensing date |
| CN109119976A (en) * | 2018-09-05 | 2019-01-01 | 国网湖南省电力有限公司 | The automatic adjusting method and system of relay protection constant value for transmission line forest fire |
| WO2020083091A1 (en) * | 2018-10-25 | 2020-04-30 | 国网湖南省电力有限公司 | Risk analysis method and system for power grid fault |
| CN109377701A (en) * | 2018-10-29 | 2019-02-22 | 国网四川省电力公司电力科学研究院 | The real-time early warning system and method for transmission line forest fire fire behavior development trend |
Non-Patent Citations (3)
| Title |
|---|
| FANGRONG ZHOU等: ""Research on transmission line fire monitoring technology based on remote sensing satellite data"", 《2020 INTERNATIONAL CONFERENCE ON URBAN ENGINEERING AND MANAGEMENT SCIENCE (ICUEMS)》 * |
| 段翔兮等: ""基于告警信息的保护跳闸识别方法研究"", 《四川电力技术》 * |
| 熊小伏等: ""基于山火时空特征的林区输电通道风险评估"", 《电力系统保护与控制》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973584A (en) * | 2022-05-10 | 2022-08-30 | 云南电网有限责任公司电力科学研究院 | Mountain fire warning method and device, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112949515B (en) | 2022-07-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102629294B (en) | Probability evaluation method of failure caused by typhoon to power transmission line | |
| CN103246936A (en) | System and method for pre-warning of typhoon risks of overhead transmission lines of grid | |
| CN102446390A (en) | Method and system for safety detection and early warning of monitoring areas near power transmission lines | |
| CN106844595A (en) | A kind of specialized meteorological services application platform based on power grid GIS | |
| CN109543870B (en) | Power transmission line tower lightning stroke early warning method based on neighborhood preserving embedding algorithm | |
| CN112949515B (en) | Line forest fire early warning method and system based on monitoring information | |
| CN114383652A (en) | Method, system and device for identifying potential fault online risk of power distribution network | |
| CN117078017A (en) | Intelligent decision analysis system for monitoring power grid equipment | |
| CN112185086B (en) | Grain depot safety early warning system based on deviation analysis | |
| CN111738617A (en) | Transformer substation risk assessment method and early warning system in heavy rainfall weather | |
| CN106296453A (en) | A kind of electric power tripping operation statistical method | |
| CN116799946A (en) | Power grid holographic intelligent management and control method and platform based on risk scheduling | |
| CN115749955A (en) | Tailing pond safety risk real-time early warning system based on multi-dimensional perception | |
| CN206075498U (en) | Electrical network windburn early warning system based on numerical simulation weather forecast | |
| CN111313355B (en) | Method for updating monitoring signal event rule under manual supervision | |
| CN108802571B (en) | Fault early warning method based on spread spectrum acquisition and tangent linear variance and implementation system thereof | |
| CN105678460B (en) | The correlation analysis method of Lightning Disaster and apparatus overload and cross section limit | |
| CN113224850A (en) | Electric power monitoring system and method | |
| CN113269509A (en) | Construction method of electric power meteorological database | |
| CN114118598A (en) | Power grid lightning disaster prediction method and system based on deep learning prediction model | |
| CN113092932B (en) | Rapid early warning method for mountain fire fault of power grid line | |
| Xu et al. | Early Warning Method of Power Grid Meteorological Disasters Based on Scene Recognition | |
| Fanqin et al. | Study on Fault Analysis and early warning system of wind-induced disasters on transmission lines | |
| CN108767858A (en) | Broken emergency processing method and system outside transmission line of electricity based on GIS platform | |
| Wang et al. | Comprehensive monitoring cycle classification of centralized monitoring substations based on knowledge map and competitive neural network |
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 | ||
| GR01 | Patent grant |