CN114386699A - Method and system for predicting operation condition of base station after earthquake - Google Patents

Method and system for predicting operation condition of base station after earthquake Download PDF

Info

Publication number
CN114386699A
CN114386699A CN202210036950.4A CN202210036950A CN114386699A CN 114386699 A CN114386699 A CN 114386699A CN 202210036950 A CN202210036950 A CN 202210036950A CN 114386699 A CN114386699 A CN 114386699A
Authority
CN
China
Prior art keywords
base station
earthquake
data
global
local
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
Application number
CN202210036950.4A
Other languages
Chinese (zh)
Other versions
CN114386699B (en
Inventor
常岐海
李学明
明政
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xizang Minzu University
Original Assignee
Xizang Minzu University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xizang Minzu University filed Critical Xizang Minzu University
Priority to CN202210036950.4A priority Critical patent/CN114386699B/en
Publication of CN114386699A publication Critical patent/CN114386699A/en
Application granted granted Critical
Publication of CN114386699B publication Critical patent/CN114386699B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/11—Complex mathematical operations for solving equations, e.g. nonlinear equations, general mathematical optimization problems
    • G06F17/12—Simultaneous equations, e.g. systems of linear equations
    • 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
    • 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
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Business, Economics & Management (AREA)
  • Operations Research (AREA)
  • Databases & Information Systems (AREA)
  • Strategic Management (AREA)
  • Algebra (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Development Economics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

The invention relates to a method and a system for predicting the operation condition of a base station after an earthquake. Belongs to the technical field of prediction of operation conditions of base stations after earthquakes. The method comprehensively analyzes the data such as the magnitude of earthquake, the earthquake intensity, the relative earthquake center position of the base station, the geological structure of the position of the base station, the geographic distribution characteristics of the position of the base station, the preset operation data of the base station and the like, and further predicts the operation condition of the base station after the earthquake, thereby avoiding the defect of single prediction parameter in the existing prediction method of the operation condition of the base station after the earthquake. The operation condition of the base station after the earthquake is predicted by utilizing a plurality of variable parameter data influencing the operation condition of the base station after the earthquake, so that the prediction accuracy can be improved.

Description

Method and system for predicting operation condition of base station after earthquake
Technical Field
The invention relates to the technical field of prediction of whether an earthquake-caused base station operates or not, in particular to a method and a system for predicting the operation condition of the earthquake-caused base station.
Background
A base station, i.e., a common mobile communication base station, is one form of a radio station. The base station comprises an iron tower part and a machine room part; the operation conditions of the base station include two kinds, i.e., operation and non-operation. If a large earthquake occurs, the base station is damaged, so that the communication service is interrupted in a large area, an island effect is formed, and difficulty is added to rescue after the earthquake. Therefore, the method can predict which base stations are still in operation and which base stations cannot operate after the earthquake, and preferentially rescue the base stations in the main places, and has great significance for communication rescue and disaster reduction after the earthquake.
Prior research has focused on the relationship between aspects of base station communications equipment, building structures, etc. and earthquakes. And predicting the failure probability of the base station through vulnerability analysis and a fault tree model of the base station. For example, in the morning, "study on earthquake vulnerability of a typical mobile communication base station", it is essential to study the earthquake-resistant effect of the iron tower of the base station, the equipment in the machine room and the building in the machine room, and predict the failure probability of the base station after the earthquake according to the earthquake-resistant effect. However, the influence of an earthquake on the base station is a complicated process, and for example, the base station may be influenced by debris flow caused by the earthquake. Therefore, the existing prediction method for the operation of the base station after the earthquake is slightly unilateral, and the prediction accuracy is not high. In order to improve the accuracy of the operation prediction of the base station after the earthquake, the invention provides a method and a system for predicting the operation condition of the base station after the earthquake.
Disclosure of Invention
The invention aims to provide a method and a system for predicting the operation condition of a base station after an earthquake. The method is used for improving the accuracy of predicting the operation condition of the base station after the earthquake.
In order to achieve the purpose, the invention provides the following scheme:
in one aspect, the present invention provides a method for predicting an operation condition of a base station after an earthquake, which specifically includes:
according to historical earthquake disaster statistical data of earthquake-prone areas, establishing a variable parameter database influencing the operation conditions of the base station after an earthquake; the variable parameter data includes: the earthquake magnitude, the earthquake intensity, the relative epicenter azimuth of the base station, the geological structure of the position of the base station, the geographical distribution characteristics of the position of the base station and the preset operation data of the base station;
preprocessing variable parameter data in a database to obtain a data set for predicting the operation condition of the base station after the earthquake;
dividing the data set into a test set and a training set according to a preset threshold value;
training a logistic regression analysis model by using the training set to obtain a trained logistic regression analysis model;
testing the trained logistic regression analysis model by using a test set, and if the test is not qualified, returning to the step of dividing the data set into the test set and the training set according to the preset threshold after adjusting the preset threshold; if the test is qualified, a final prediction model is obtained;
and predicting the operation condition of the base station after the earthquake by using the final prediction model.
Optionally, the method for predicting the operation condition of the base station after the earthquake further comprises the step of forming the variable parameter data in the variable parameter data database into one-dimensional data so as to improve the prediction speed.
Optionally, the geological structure of the position of the base station includes sedimentary rock, magmatic rock and metamorphic rock;
the geographical distribution characteristics of the positions of the base stations comprise the positions and the properties of the base stations;
the base station position comprises a mountain top base station, a semi-hillside base station and a flat ground base station;
the base station properties include a ground base station and a floor base station.
Optionally, the preprocessing the variable parameter data in the database includes:
carrying out collinearity detection on the variable parameter data to obtain variable parameter data without collinearity;
and carrying out abnormal value processing on the variable parameter data without collinearity to obtain a data set for predicting the operation condition of the base station after the earthquake.
Optionally, the performing colinearity detection on the variable parameter data includes:
calculating a variance expansion factor of each variable parameter data, comparing the variance expansion factor with a first threshold value, and if the variance expansion factor is greater than or equal to the first threshold value, rejecting the variable parameter data; and if the variance expansion factor is smaller than the first threshold value, the variable parameter data is reserved to obtain variable parameter data without collinearity.
Optionally, the training of the logistic regression analysis model by using the training set to obtain the trained logistic regression analysis model includes:
training a logistic regression analysis model by using the training set to obtain an initial training model;
respectively carrying out Omnibus inspection, model abstract inspection and Hosimer-Lemeixiao inspection on the model coefficients of the initial training model, and obtaining a trained logistic regression analysis model if the three inspections are passed; otherwise, returning to the step of preprocessing the variable parameter data in the database.
Optionally, the predicting the operation condition of the post-earthquake base station by using the final prediction model includes:
inputting the variable parameter data into the final prediction model to obtain a prediction probability;
comparing the prediction probability with a preset critical value, and if the prediction probability is greater than or equal to the preset critical value, determining that the operation condition of the base station is operation; and if the prediction probability is smaller than the preset critical value, the base station is not operated under the operation working condition.
Optionally, the preset operation data of the base station includes whether the machine room seismic performance is qualified or not and whether the iron tower seismic performance is qualified or not;
and respectively judging whether the anti-seismic performance of the machine room is qualified or not and whether the anti-seismic performance of the iron tower is qualified or not according to the earthquake level fortification data, the earthquake level preset data, the earthquake intensity fortification data and the earthquake intensity preset data.
Optionally, the method for predicting the operation condition of the post-earthquake base station provided by the invention further includes:
grouping variable parameter data in a database to obtain local variable parameter data and global variable parameter data; the local variable parameter data comprises earthquake magnitude, earthquake intensity, relative epicenter azimuth of the base station and preset operation data of the base station; the global variable parameter data comprise earthquake magnitude, earthquake intensity, a geological structure of the position of the base station and geographic distribution characteristics of the position of the base station;
preprocessing the local variable parameter data to obtain a local data set; preprocessing the global variable parameter data to obtain a global data set;
dividing the local data set into a local test set and a local training set according to a preset threshold value;
training a local regression analysis model by using a local training set to obtain a trained local logistic regression analysis model;
checking the trained local logistic regression analysis model by using a local test set, if the test is unqualified, adjusting the preset threshold value and returning to the step of dividing the local data set into a local test set and a local training set according to the preset threshold value; if the test is qualified, obtaining a final local prediction model;
predicting the operation condition of the base station after the earthquake by using the local final prediction model to obtain local prediction probability;
dividing the global data set into a global test set and a global training set according to a preset threshold value;
training a logistic regression analysis model by using the global training set to obtain a trained global logistic regression analysis model;
checking the trained global logistic regression analysis model by using a global test set, if the check is not qualified, adjusting the preset threshold value and returning to the step of dividing the global data set into the global test set and the global training set according to the preset threshold value; if the global final prediction model is qualified, obtaining a global final prediction model;
predicting the operation condition of the base station after the earthquake by using the global final prediction model to obtain global prediction probability;
and predicting the operation condition of the base station after the earthquake according to the local prediction probability and the global prediction probability.
Optionally, the predicting, according to the local prediction probability and the global prediction probability, the operation condition of the post-earthquake base station includes:
acquiring a plurality of groups of local prediction probabilities and global prediction probabilities;
weighting the local prediction probability and the global prediction probability in each group, and calculating a weight parameter by using a main value analysis method;
obtaining the prediction probability of the operation condition of the base station after the earthquake according to the weight parameters, the local prediction probability and the global prediction probability;
comparing the prediction probability with a preset critical value, and if the prediction probability is greater than or equal to the preset critical value, determining that the operation condition of the base station is operation; and if the prediction probability is smaller than the preset critical value, the base station is not operated under the operation working condition.
On the other hand, the invention also provides a system for predicting the operation condition of the base station after the earthquake, which comprises the following steps:
the data acquisition module is used for establishing a variable parameter database influencing the operation condition of the base station after the earthquake according to historical earthquake disaster statistical data of the earthquake-prone area; the variable parameter data includes: the earthquake magnitude, the earthquake intensity, the relative epicenter azimuth of the base station, the geological structure of the position of the base station, the geographical distribution characteristics of the position of the base station and the preset operation data of the base station;
the data preprocessing module is used for preprocessing the variable parameter data in the database to obtain a data set used for predicting the operation condition of the base station after the earthquake;
the data dividing module is used for dividing the data set into a test set and a training set according to a preset threshold value;
the model training module is used for training the logistic regression analysis model by using the training set to obtain a trained logistic regression analysis model;
the model testing module is used for testing the trained logistic regression analysis model by using a test set, and if the test is unqualified, the step of dividing the data set into the test set and the training set according to the preset threshold value is returned after the preset threshold value is adjusted; if the test is qualified, a final prediction model is obtained;
and the data prediction module is used for predicting the operation condition of the base station after the earthquake by using the final prediction model.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the invention discloses a method and a system for predicting the operation condition of a base station after an earthquake, wherein the method specifically comprises the following steps: according to historical earthquake disaster statistical data of earthquake-prone areas, establishing a variable parameter database influencing the operation conditions of the base station after an earthquake; the variable parameter data includes: the earthquake magnitude, the earthquake intensity, the relative epicenter azimuth of the base station, the geological structure of the position of the base station, the geographical distribution characteristics of the position of the base station and the preset operation data of the base station; preprocessing variable parameter data in a database to obtain a data set for predicting the operation condition of the base station after the earthquake; dividing the data set into a test set and a training set according to a preset threshold value; training a logistic regression analysis model by using the training set to obtain a trained logistic regression analysis model; testing the trained logistic regression analysis model by using a test set, and if the test is not qualified, returning to the step of dividing the data set into the test set and the training set according to the preset threshold after adjusting the preset threshold; if the test is qualified, a final prediction model is obtained; and predicting the operation condition of the base station after the earthquake by using the final prediction model.
(1) After an earthquake occurs, the invention comprehensively analyzes data such as the magnitude of the earthquake, the intensity of the earthquake, the relative epicenter position of the base station, the geological structure of the position of the base station, the geographic distribution characteristics of the position of the base station, the preset operation data of the base station and the like, and further predicts the operation condition of the base station after the earthquake, thereby avoiding the defect of single prediction parameter in the existing prediction method of the operation condition of the base station after the earthquake. The operation condition of the base station after the earthquake is predicted by utilizing a plurality of variable parameter data influencing the operation condition of the base station after the earthquake, so that the prediction accuracy can be improved. Is beneficial to emergency rescue and disaster relief after earthquake.
(2) The invention can be used for evaluating the operation of the base station and evaluating the site selection of the base station construction.
(3) The final predictive model determined by the above method of the present invention may be used to evaluate the base station operation of any seismic source.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for predicting the operation condition of a base station after an earthquake according to the present invention;
FIG. 2 is a residual error diagram in data preprocessing of the method for predicting the operation condition of the base station after an earthquake according to the present invention;
FIG. 3 is a residual error diagram of residual error value removal in data preprocessing of the method for predicting the operation condition of the base station after an earthquake according to the present invention;
FIG. 4 is a diagram illustrating a prediction result of a method for predicting an operation condition of a base station after an earthquake according to the present invention, wherein 0.5 is a preset critical value;
FIG. 5 is a diagram illustrating a prediction result of a method for predicting an operation condition of a base station after an earthquake according to the present invention, wherein 0.4 is a predetermined critical value;
FIG. 6 is a schematic diagram of a system for predicting the operation condition of a base station after an earthquake according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
The existing technology for predicting the operation condition of the base station after the earthquake mainly focuses on the relation between the communication equipment of the base station and the earthquake or the relation between a machine room building structure and the earthquake. Namely, the operation condition of the base station after the earthquake is predicted according to the vulnerability of the base station equipment. The method is essentially used for researching the anti-seismic effect of the base station iron tower, the machine room equipment and the base station building and predicting the failure probability of the base station after the earthquake according to the anti-seismic effect of the base station iron tower, the machine room equipment and the base station building. The prediction method is relatively simple and does not fully consider factors influencing the operation condition of the base station after the earthquake. Therefore, the prediction result is not accurate. Based on the method, the invention provides a method and a system for predicting the operation condition of a base station after an earthquake. The method is used for improving the accuracy of the prediction of the operation condition of the base station after the earthquake.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
Example 1
As shown in fig. 1, the present embodiment provides a method for predicting an operation condition of a base station after an earthquake, which includes the following specific steps:
s1: according to historical earthquake disaster statistical data of earthquake-prone areas, establishing a variable parameter database influencing the operation conditions of the base station after an earthquake; the variable parameter data includes: the earthquake magnitude, the earthquake intensity, the relative epicenter azimuth of the base station, the geological structure of the position of the base station, the geographical distribution characteristics of the position of the base station and the preset operation data of the base station.
Preferably, in order to prevent the data of the base station after earthquake in a single area from being too comprehensive, the statistical data of earthquake disasters can adopt data after 8.0-level great earthquake in Wenchuan, Lushan 7.0-level earthquake and 6.0-level Changning earthquake in Sichuan province.
Preferably, the geological structure of the base station comprises sedimentary rock, magmatic rock and metamorphic rock; the geographical distribution characteristics of the positions of the base stations comprise the positions and the properties of the base stations; the base station position comprises a mountain top base station, a semi-hillside base station and a flat ground base station; the base station properties comprise a ground base station and a floor base station; the preset operation data of the base station comprise whether the machine room seismic performance is qualified or not and whether the iron tower seismic performance is qualified or not.
A person skilled in the art can respectively judge whether the anti-seismic performance of the machine room is qualified or not and whether the anti-seismic performance of the iron tower is qualified or not according to the earthquake level fortification data, the earthquake level preset data, the earthquake intensity fortification data and the earthquake intensity preset data. The fortification data refers to fortification values determined according to relevant national regulations in base station construction. The preset data refers to parameter values which are manually set according to requirements in the prediction of the operation condition of the base station after the earthquake. In specific implementation, the logic of the earthquake resistance of the machine room is k, the logic of the earthquake resistance of the iron tower is H, and the weight parameters of the earthquake intensity and the earthquake magnitude of the machine room are respectively set to be m1And m2Seismic intensity and shock of iron towerThe weight parameters of the stages are respectively set to n1And n2The seismic intensity and magnitude are set as alpha and beta, respectively, and e is a constant. Each machine room or iron tower corresponds to a binary linear equation respectively, and a binary linear equation set is established according to seismic intensity, seismic fortification data of seismic level and data of preset seismic intensity and preset seismic level.
The system of linear equations of the computer room is as follows:
Figure BDA0003468282550000081
wherein k isFortificationRepresenting the seismic fortification value, k, of the machine roomPreset ofIndicating the preset value of the machine room, alphaFortificationRepresenting seismic intensity fortification data, m1Weight parameter, beta, representing seismic intensityFortificationRepresenting seismic fortification data, m2Weight parameter representing magnitude, e1Is a constant, αPreset ofSeismic intensity preset data, betaPreset ofRepresenting preset data of magnitude, e2Is a constant. M can be calculated by principal value analysis1、m2、e1、e2The value of (c).
In practical application, the machine room earthquake fortification value is compared with the machine room earthquake preset value, if the machine room earthquake fortification value is larger than the machine room earthquake preset value, the machine room earthquake resistance is qualified, and otherwise, the machine room earthquake fortification value is unqualified.
The system of the two-dimensional linear equations of the iron tower is as follows:
Figure BDA0003468282550000082
wherein HFortificationRepresenting the seismic fortification value of the iron tower, HPreset ofRepresenting the predetermined value of the tower resistance to vibration, alphaFortificationRepresenting seismic intensity fortification data, n1Weight parameter, beta, representing seismic intensityFortificationRepresenting seismic fortification data, n2Weight parameter representing magnitude, e3Is a constant, αPreset ofSeismic intensity preset data, betaPreset ofRepresenting preset data of magnitude, e4Is a constant. N can be calculated by principal value analysis1、n2、e3、e4The value of (c).
In practical application, the steel tower earthquake fortification value is compared with a steel tower earthquake preset value, if the steel tower earthquake fortification value is larger than the steel tower earthquake preset value, the steel tower earthquake performance is qualified, and otherwise, the steel tower earthquake fortification value is unqualified.
Whether the anti-seismic performance of the machine room is qualified or not and whether the anti-seismic performance of the iron tower is qualified or not are judged through the formula, so that the prediction accuracy of the operation working condition of the base station after the earthquake can be improved.
In specific implementation, in order to improve the prediction speed, the variable parameter data in the variable parameter data database may be further combined into one-dimensional data. The form of the one-dimensional data includes: and the weight of the northeast (8-grade earthquake, 10-degree intensity, granite geological quality, unqualified earthquake resistance of a machine room, a base station in a semi-hillside, a floor base station and qualified earthquake resistance of an iron tower) is g, wherein the weight g is related to the operation condition of the base station after the earthquake.
S2: and preprocessing the variable parameter data in the database to obtain a data set for predicting the operation condition of the base station after the earthquake.
The specific steps of preprocessing the variable parameter data in the database comprise:
s21: and carrying out co-linearity detection on the variable parameter data to obtain variable parameter data without co-linearity.
The collinearity means that each variable parameter in the variable parameter data can be determined by the rest variable parameters, and each variable parameter can also determine the rest variable parameters, if the collinearity phenomenon appears in the data, the inference and prediction in the regression analysis can be seriously influenced. Therefore, co-linear detection of variable parameter data is required. The specific detection method comprises the following steps:
calculating a variance expansion factor of each variable parameter data, comparing the variance expansion factor with a first threshold value, and if the variance expansion factor is greater than or equal to the first threshold value, rejecting the variable parameter data; and if the variance expansion factor is smaller than the first threshold value, the variable parameter data is reserved to obtain variable parameter data without collinearity.
The Variance Inflation Factor (VIF) is calculated by the following formula:
Figure BDA0003468282550000091
in the formula, R2The goodness-of-fit index is also called a decision coefficient, and is the ratio of the regression sum of squares to the total sum of squared deviations, and i represents the variable parameter data. Specifically, when 0<VIF<When the variable parameter data is less than or equal to 10, the variable parameter data has no collinearity problem, and when the variable parameter data is less than or equal to 10, the variable parameter data has collinearity. The step-by-step regression method can be used for carrying out rejecting operation on the variable parameter data with the collinearity; the variable parameter data with collinearity can be manually removed by carrying out correlation analysis on the input variable parameter data; the influence caused by the co-linear variable parameter data can be reduced by expanding the quantity of the variable parameter data.
Specifically, the VIF values for the variable parameters are shown in table 1:
TABLE 1 VIF values
Figure BDA0003468282550000092
As can be seen from the table above, the VIF values among the variable parameters of the invention are all less than 2 according to the formula calculation result, and the collinearity detection standard of the variance expansion factor VIF is met, so that the collinearity problem does not exist in the variable parameter data which is selected as the factors influencing the operation of the base station.
S22: and carrying out abnormal value processing on the variable parameter data without collinearity to obtain a data set for predicting the operation condition of the base station after the earthquake. The method for processing the abnormal value comprises the following steps:
and calculating a standardized residual value of the variable parameter data, taking the fitting value of the variable parameter data as an abscissa of a two-dimensional coordinate system, and taking the standardized residual value as an ordinate of the two-dimensional coordinate system to obtain a residual map, wherein the residual map is shown as 2.
And judging whether the variable parameter data is an abnormal value or not according to the residual map, if the standardized residual value of the variable parameter data is within a preset interval, judging that the variable parameter data is not abnormal, otherwise, judging that the variable parameter data is the abnormal value and needing to be removed so as to obtain a data set for predicting the operation condition of the base station after the earthquake. Preferably, the value range of the preset interval is (-2, 2).
Specifically, the calculation formula of the normalized residual value is as follows:
Figure BDA0003468282550000101
in the formula, eiIs a common residual error, and the residual error is,
Figure BDA0003468282550000102
is the mean square error, i represents the variable parameter data, PiiAre the elements on the diagonal of the matrix P.
As can be seen from fig. 2, there are two sets of abnormal values, which need to be rejected, and the result after rejection is shown in fig. 3.
S3: and dividing the data set into a test set and a training set according to a preset threshold value.
In particular implementations, 70% of the data in the data set may be divided into a training set, with the remaining 30% of the data being used as a test set.
S4: and training a logistic regression analysis model by using the training set to obtain the trained logistic regression analysis model.
Specifically, the logistic regression analysis model is a binary logistic model, and the expression of the logistic regression analysis model is as follows:
Figure BDA0003468282550000103
wherein, X1...XiIs the input variable parameter data, Y is the output variable, which is related to whether the base station operates after earthquake, beta0...βiIs a rightParameter, P is probability.
The specific steps for obtaining the trained logistic regression analysis model include:
s41: and training a logistic regression analysis model by using the training set to obtain an initial training model.
S42: respectively carrying out Omnibus inspection, model abstract inspection and Hosimer-Lemeixiao inspection on the model coefficients of the initial training model, and obtaining a trained logistic regression analysis model if the three inspections are passed; otherwise, the process returns to step S2.
Specifically, in the Omnibus test of the model coefficient, whether the significance (P value) of the model is less than 0.05 or not is judged, if so, the test is passed, and if not, the test is not passed; in the model abstract inspection, judging whether the-2 log likelihood value is larger than a first preset threshold value, if so, passing the inspection, otherwise, failing the inspection; in the Hosimer-Leimei test, whether the significance (P value) in the model coefficient is greater than 0.05 or not is judged, if so, the test is passed, and if not, the test is not passed.
S5: checking the trained logistic regression analysis model by using a test set, and if the checking is unqualified, adjusting the preset threshold value and returning to the step S'; and if the test is qualified, obtaining a final prediction model.
S6: and predicting the operation condition of the base station after the earthquake by using the final prediction model.
Preferably, the predicting the operation condition of the post-earthquake base station by using the final prediction model comprises:
s61: and inputting the variable parameter data into the final prediction model to obtain the prediction probability.
S62: comparing the prediction probability with a preset critical value, and if the prediction probability is greater than or equal to the preset critical value, determining that the operation condition of the base station is operation; and if the prediction probability is smaller than the preset critical value, the base station is not operated under the operation working condition.
In specific implementation, the preset critical value can be adjusted to improve the accuracy of the operation condition of the base station after the earthquake. Taking fig. 4 as an example, where the preset critical value is 0.5, the model training set predicts 19 successful base stations and 2 failed base stations, and the prediction accuracy is 90.5%. The model test set successfully predicts 19 base stations and fails to predict 2 base stations, and the prediction accuracy is 90.5%. In order to improve the prediction accuracy, the preset critical value may be adjusted to 0.4, as shown in fig. 5, the model training set predicts 56 successful base stations and 3 failed base stations, and the prediction success rate is 94.9%, which is 1.7% higher than the original one. The model test set predicts 20 successful base stations and 1 failed base station, the prediction success rate is 95.2%, and the prediction accuracy is improved by 4.7%.
Because actual disaster statistical data is often less than a given multiple standard of variable parameters, in order to improve prediction accuracy, the variable parameter data in the database can be grouped, the operation condition prediction of the base station after the earthquake is carried out on each group of variable parameter data respectively to obtain a plurality of groups of prediction probabilities, and then a final prediction result is obtained according to the plurality of groups of prediction probabilities. The specific treatment process is as follows:
grouping the variable parameter data in the database to obtain local variable parameter data and global variable parameter data; the local variable parameter data comprises earthquake magnitude, earthquake intensity, relative epicenter azimuth of the base station and preset operation data of the base station; the global variable parameter data comprises earthquake magnitude, earthquake intensity, geological structure of the position of the base station and geographical distribution characteristics of the position of the base station.
Preprocessing the local variable parameter data to obtain a local data set; and preprocessing the global variable parameter data to obtain a global data set.
And dividing the local data set into a local test set and a local training set according to a preset threshold value.
And training the local regression analysis model by using a local training set to obtain the trained local logistic regression analysis model.
Checking the trained local logistic regression analysis model by using a local test set, if the test is unqualified, adjusting the preset threshold value and returning to the step of dividing the local data set into a local test set and a local training set according to the preset threshold value; and if the test is qualified, obtaining a local final prediction model.
And predicting the operation condition of the base station after the earthquake by using the local final prediction model to obtain local prediction probability.
And dividing the global data set into a global test set and a global training set according to a preset threshold value.
And training a logistic regression analysis model by using the global training set to obtain the trained global logistic regression analysis model.
Checking the trained global logistic regression analysis model by using a global test set, if the check is not qualified, adjusting the preset threshold value and returning to the step of dividing the global data set into the global test set and the global training set according to the preset threshold value; and if the global final prediction model is qualified, obtaining a global final prediction model.
And predicting the operation condition of the base station after the earthquake by using the global final prediction model to obtain global prediction probability.
And predicting the operation condition of the base station after the earthquake according to the local prediction probability and the global prediction probability.
Preferably, the predicting the operation condition of the post-earthquake base station according to the local prediction probability and the global prediction probability includes:
and acquiring a plurality of groups of local prediction probabilities and global prediction probabilities.
And weighting the local prediction probability and the global prediction probability in each group, and calculating a weight parameter by using a main value analysis method.
And obtaining the prediction probability of the operation condition of the base station after the earthquake according to the weight parameter, the local prediction probability and the global prediction probability.
Comparing the prediction probability with a preset critical value, and if the prediction probability is greater than or equal to the preset critical value, determining that the operation condition of the base station is operation; and if the prediction probability is smaller than the preset critical value, the base station is not operated under the operation working condition.
The method solves the problem that the prediction factor is single in the existing method for predicting the operation condition of the base station after the earthquake, and can more accurately predict the operation condition of the base station after the earthquake. Meanwhile, the existing method for predicting the operation condition of the base station after the earthquake utilizes a fault tree model to predict the operation condition of the base station, and the data source of the fault tree model and a sensor arranged on the site are likely to be damaged by strong earthquake waves when the earthquake occurs. Therefore, its data source is not stable. Instability of the data source may further affect the accuracy of the prediction results. The method utilizes a logistic regression analysis model to predict the operation condition of the base station after the earthquake, and the data of the method is derived from historical earthquake disaster statistical data of earthquake-prone areas. The data source is stable, and the prediction result is more accurate. The method uses the logistic regression analysis model to process the variable parameter data to predict the operation condition of the base station after the earthquake, and can obtain the significance of each variable parameter data. The result shows that the significance of the variable parameter data conforms to the specified threshold value, so that the method has rationality for predicting the operation condition of the base station after the earthquake by using the variable parameter data. The significance of the variable parameter data is sorted from big to small to obtain the seismic intensity, the seismic level, whether the machine room seismic performance is qualified, whether the iron tower seismic performance is qualified, the geological structure, the base station property, the base station position and the relative seismic center position of the base station.
In specific implementation, the method can also be used for analyzing the site selection of the base station construction.
Example 2
As shown in fig. 6, this embodiment provides a system for predicting an operation condition of a base station after an earthquake by applying the method provided in embodiment 1, which specifically includes:
the data acquisition module is used for establishing a variable parameter database influencing the operation condition of the base station after the earthquake according to historical earthquake disaster statistical data of the earthquake-prone area; the variable parameter data includes: the earthquake magnitude, the earthquake intensity, the relative epicenter azimuth of the base station, the geological structure of the position of the base station, the geographical distribution characteristics of the position of the base station and the preset operation data of the base station.
And the data preprocessing module is used for preprocessing the variable parameter data in the database to obtain a data set for predicting the operation condition of the base station after the earthquake.
And the data dividing module is used for dividing the data set into a test set and a training set according to a preset threshold value.
And the model training module is used for training the logistic regression analysis model by using the training set to obtain the trained logistic regression analysis model.
The model testing module is used for testing the trained logistic regression analysis model by using a test set, and if the test is unqualified, the step of dividing the data set into the test set and the training set according to the preset threshold value is returned after the preset threshold value is adjusted; and if the test is qualified, obtaining a final prediction model.
And the data prediction module is used for predicting the operation condition of the base station after the earthquake by using the final prediction model.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims (6)

1. A method for predicting the operating conditions of a post-earthquake base station, comprising:
according to historical earthquake disaster statistical data of earthquake-prone areas, establishing a variable parameter database influencing the operation conditions of the base station after an earthquake; the variable parameter data includes: the earthquake magnitude, the earthquake intensity, the relative epicenter azimuth of the base station, the geological structure of the position of the base station, the geographical distribution characteristics of the position of the base station and the preset operation data of the base station;
preprocessing variable parameter data in a database to obtain a data set for predicting the operation condition of the base station after the earthquake;
dividing the data set into a test set and a training set according to a preset threshold value;
training a logistic regression analysis model by using the training set to obtain a trained logistic regression analysis model;
testing the trained logistic regression analysis model by using a test set, and if the test is not qualified, returning to the step of dividing the data set into the test set and the training set according to the preset threshold after adjusting the preset threshold; if the test is qualified, a final prediction model is obtained;
and predicting the operation condition of the base station after the earthquake by using the final prediction model.
2. The method of predicting the operating conditions of a post-seismic base station as set forth in claim 1, further comprising:
grouping variable parameter data in a database to obtain local variable parameter data and global variable parameter data; the local variable parameter data comprises earthquake magnitude, earthquake intensity, relative epicenter azimuth of the base station and preset operation data of the base station; the global variable parameter data comprise earthquake magnitude, earthquake intensity, a geological structure of the position of the base station and geographic distribution characteristics of the position of the base station;
preprocessing the local variable parameter data to obtain a local data set; preprocessing the global variable parameter data to obtain a global data set;
dividing the local data set into a local test set and a local training set according to a preset threshold value;
training a local regression analysis model by using a local training set to obtain a trained local logistic regression analysis model;
checking the trained local logistic regression analysis model by using a local test set, if the test is unqualified, adjusting the preset threshold value and returning to the step of dividing the local data set into a local test set and a local training set according to the preset threshold value; if the test is qualified, obtaining a final local prediction model;
predicting the operation condition of the base station after the earthquake by using the local final prediction model to obtain local prediction probability;
dividing the global data set into a global test set and a global training set according to a preset threshold value;
training a logistic regression analysis model by using the global training set to obtain a trained global logistic regression analysis model;
checking the trained global logistic regression analysis model by using a global test set, if the check is not qualified, adjusting the preset threshold value and returning to the step of dividing the global data set into the global test set and the global training set according to the preset threshold value; if the global final prediction model is qualified, obtaining a global final prediction model;
predicting the operation condition of the base station after the earthquake by using the global final prediction model to obtain global prediction probability;
and predicting the operation condition of the base station after the earthquake according to the local prediction probability and the global prediction probability.
3. The method of claim 2, wherein the predicting the operation condition of the post-earthquake base station according to the local prediction probability and the global prediction probability comprises:
acquiring a plurality of groups of local prediction probabilities and global prediction probabilities;
weighting the local prediction probability and the global prediction probability in each group, and calculating a weight parameter by using a main value analysis method;
obtaining the prediction probability of the operation condition of the base station after the earthquake according to the weight parameters, the local prediction probability and the global prediction probability;
comparing the prediction probability with a preset critical value, and if the prediction probability is greater than or equal to the preset critical value, determining that the operation condition of the base station is operation; and if the prediction probability is smaller than the preset critical value, the base station is not operated under the operation working condition.
4. The method for predicting the operation condition of the post-earthquake base station according to claim 1, wherein the step of predicting the operation condition of the post-earthquake base station by using the final prediction model comprises the following steps:
inputting the variable parameter data into the final prediction model to obtain a prediction probability;
comparing the prediction probability with a preset critical value, and if the prediction probability is greater than or equal to the preset critical value, determining that the operation condition of the base station is operation; and if the prediction probability is smaller than the preset critical value, the base station is not operated under the operation working condition.
5. The method for predicting the operating condition of the post-earthquake base station according to claim 1, wherein the preset operating data of the base station comprise whether the earthquake resistance of the machine room is qualified or not and whether the earthquake resistance of the iron tower is qualified or not;
and respectively judging whether the anti-seismic performance of the machine room is qualified or not and whether the anti-seismic performance of the iron tower is qualified or not according to the earthquake level fortification data, the earthquake level preset data, the earthquake intensity fortification data and the earthquake intensity preset data.
6. A system for predicting the operating conditions of a post-earthquake base station, comprising:
the data acquisition module is used for establishing a variable parameter database influencing the operation condition of the base station after the earthquake according to historical earthquake disaster statistical data of the earthquake-prone area; the variable parameter data includes: the earthquake magnitude, the earthquake intensity, the relative epicenter azimuth of the base station, the geological structure of the position of the base station, the geographical distribution characteristics of the position of the base station and the preset operation data of the base station;
the data preprocessing module is used for preprocessing the variable parameter data in the database to obtain a data set used for predicting the operation condition of the base station after the earthquake;
the data dividing module is used for dividing the data set into a test set and a training set according to a preset threshold value;
the model training module is used for training the logistic regression analysis model by using the training set to obtain a trained logistic regression analysis model;
the model testing module is used for testing the trained logistic regression analysis model by using a test set, and if the test is unqualified, the step of dividing the data set into the test set and the training set according to the preset threshold value is returned after the preset threshold value is adjusted; if the test is qualified, a final prediction model is obtained;
and the data prediction module is used for predicting the operation condition of the base station after the earthquake by using the final prediction model.
CN202210036950.4A 2022-01-13 2022-01-13 Method and system for predicting operation condition of base station after earthquake Active CN114386699B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210036950.4A CN114386699B (en) 2022-01-13 2022-01-13 Method and system for predicting operation condition of base station after earthquake

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210036950.4A CN114386699B (en) 2022-01-13 2022-01-13 Method and system for predicting operation condition of base station after earthquake

Publications (2)

Publication Number Publication Date
CN114386699A true CN114386699A (en) 2022-04-22
CN114386699B CN114386699B (en) 2023-06-30

Family

ID=81201442

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210036950.4A Active CN114386699B (en) 2022-01-13 2022-01-13 Method and system for predicting operation condition of base station after earthquake

Country Status (1)

Country Link
CN (1) CN114386699B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006343578A (en) * 2005-06-09 2006-12-21 Shimizu Corp Earthquake damage prediction system
CN106651039A (en) * 2016-12-27 2017-05-10 中国地震局工程力学研究所 City building group seismic hazard matrix dynamic prediction method
CN112001565A (en) * 2020-09-08 2020-11-27 清华大学合肥公共安全研究院 Earthquake disaster loss prediction and evaluation method and system based on Softmax regression model
CN112200399A (en) * 2020-07-20 2021-01-08 杭州叙简科技股份有限公司 Earthquake disaster risk assessment and economic loss prediction method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006343578A (en) * 2005-06-09 2006-12-21 Shimizu Corp Earthquake damage prediction system
CN106651039A (en) * 2016-12-27 2017-05-10 中国地震局工程力学研究所 City building group seismic hazard matrix dynamic prediction method
CN112200399A (en) * 2020-07-20 2021-01-08 杭州叙简科技股份有限公司 Earthquake disaster risk assessment and economic loss prediction method
CN112001565A (en) * 2020-09-08 2020-11-27 清华大学合肥公共安全研究院 Earthquake disaster loss prediction and evaluation method and system based on Softmax regression model

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘诗语: "通信设备及塔架地震破坏易损性分析" *

Also Published As

Publication number Publication date
CN114386699B (en) 2023-06-30

Similar Documents

Publication Publication Date Title
Castaldo et al. Seismic reliability‐based ductility demand for hardening and softening structures isolated by friction pendulum bearings
WO2022242435A1 (en) Rapid evaluation method of site seismic liquefaction disaster based on artificial intelligence
Eads et al. An efficient method for estimating the collapse risk of structures in seismic regions
CN114792020B (en) Quick evaluation method and system for building anti-seismic toughness based on machine learning
CN115203993A (en) Method, system, equipment and medium for evaluating seismic toughness of river overhead vertical frame wharf in high-intensity region
Chang et al. An empirical approach of accounting for the amplification effects induced by near-fault directivity
CN116150861A (en) Intelligent processing system for construction data of shear wall in high-intensity area
CN114386699A (en) Method and system for predicting operation condition of base station after earthquake
CN117763413A (en) Single building earthquake damage rapid assessment method based on risk census data
KR102296077B1 (en) Method of damage diagnosis of environmental basic equipment according to disaster type and the system
CN116992522A (en) Deep foundation pit support structure deformation prediction method, device, equipment and storage medium
CN115577619A (en) Method for predicting residual shear strength of liquefied soil based on machine learning
Ruggieri et al. Using transfer learning technique to define seismic vulnerability of existing buildings through mechanical models
Nabian et al. Accelerating stochastic assessment of post-earthquake transportation network connectivity via machine-learning-based surrogates
CN113516160A (en) Transmission tower attitude prediction method and system based on support vector machine
Hwang et al. Proposed methodology for earthquake-induced loss assessment of instrumented steel frame buildings: building-specific and city-scale approaches
Rota et al. Direct derivation of fragility curves from Italian post-earthquake survey data
CN113886528B (en) Method for evaluating state of overhead line after typhoon disaster based on unmanned aerial vehicle
Yan Magnitude and frequency of floods for rural, unregulated streams of Tennessee by L-Moments method
Ademović et al. SEISMIC RISK IN BOSNIA AND HERZEGOVINA BASED ON BUILDINGS’VULNERABILITY
Pejovic et al. Probabilistic Seismic Loss Assessment of RC High-Rise Buildings in Southern Euro-Mediterranean Zone
Yıldızlı Probabilistic assessment of the contribution of geotechnical factors on observed structural damage in Adapazarı and Düzce regions
Xie et al. A conception of casualty control based seismic design for buildings
Mastroberti et al. New seismic risk index for existing buildings
Papazachos et al. Results of a retrospective prediction of past strong mainshocks in the broader Aegean area by application of the accelerating seismic deformation method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant