CN111504392A - Landslide multi-element three-dimensional space monitoring system and method - Google Patents
Landslide multi-element three-dimensional space monitoring system and method Download PDFInfo
- Publication number
- CN111504392A CN111504392A CN202010521577.2A CN202010521577A CN111504392A CN 111504392 A CN111504392 A CN 111504392A CN 202010521577 A CN202010521577 A CN 202010521577A CN 111504392 A CN111504392 A CN 111504392A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- dimensional space
- landslide
- data
- positioning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention discloses a landslide multi-element three-dimensional space monitoring system and a method, wherein the system comprises the following components: the system comprises a monitoring processor, a three-dimensional space positioning monitoring unit and a multi-element sensing monitoring unit, wherein the three-dimensional space positioning monitoring unit and the multi-element sensing monitoring unit are in communication connection with the monitoring processor, and the three-dimensional space positioning monitoring unit comprises a GNSS positioning device and is used for acquiring three-dimensional positioning information and elevation information of a landslide and transmitting the three-dimensional positioning information and the elevation information to the monitoring processor; the multi-element sensing monitoring unit comprises a plurality of monitoring sensors and is used for collecting a plurality of element data reflecting the landslide state and transmitting the element data to the monitoring processor. The landslide multi-element three-dimensional space monitoring system and method organically integrate a three-dimensional space positioning technology and a multi-element real-time monitoring technology, comprehensively consider from the aspects of three-dimension and multi-dimension, realize all-dimensional, multi-element and high-precision monitoring on landslide disaster bodies, and improve the reliability, stability and effectiveness of landslide disaster monitoring.
Description
Technical Field
The invention relates to the technical field of geological disaster monitoring, in particular to a landslide multi-element three-dimensional space monitoring system and method.
Background
At present, high-precision continuous monitoring is an essential key link for landslide early warning, disaster reduction and prevention, comprises monitoring devices and technologies such as GNSS, InSAR, ground and underground, provides a key technical means for time-space domain dynamic monitoring of landslide, and particularly, the GNSS technology is widely tried to be applied to ground surface real-time deformation monitoring of landslide bodies with the advantages of all weather, high precision, easiness in construction and the like, and is expected to become one of basic guarantee technologies for landslide disaster monitoring in the future. The GNSS earth surface deformation monitoring technology mainly refers to that a global satellite positioning navigation system is utilized to realize positioning of earth surface positions of disaster points, so that once the positions of the GNSS receivers of the landslide disaster points change, the position change of the landslide disaster points can be accurately calculated through the position change between the GNSS receivers of the landslide disaster points.
Through investigation and analysis of landslide monitoring technologies at home and abroad, GNSS (global navigation satellite system) ground surface deformation monitoring at present is only used for singly acquiring three-dimensional deformation information of the ground surface of a landslide disaster body, deep association and linkage with other monitoring parameters (rainfall, water content, crack displacement, deep displacement and the like) of the landslide disaster body are not carried out, and good cross linkage is not realized in the aspect of data fusion.
Disclosure of Invention
The invention aims to provide a landslide multi-element three-dimensional space monitoring system and a landslide multi-element three-dimensional space monitoring method, which organically integrate a three-dimensional space positioning technology and a multi-element real-time monitoring technology, comprehensively consider from the aspects of three-dimension and multi-dimension, realize all-dimensional, multi-element and high-precision monitoring on a landslide disaster body, and improve the reliability, stability and effectiveness of landslide disaster monitoring.
In order to achieve the purpose, the invention provides the following scheme:
a landslide multi-element three-dimensional spatial monitoring system, the system comprising: the system comprises a monitoring processor, a three-dimensional space positioning monitoring unit and a multi-element sensing monitoring unit, wherein the three-dimensional space positioning monitoring unit and the multi-element sensing monitoring unit are in communication connection with the monitoring processor, and the three-dimensional space positioning monitoring unit comprises a GNSS positioning device and is used for acquiring three-dimensional positioning information and elevation information of a landslide and transmitting the three-dimensional positioning information and the elevation information to the monitoring processor; the multi-element sensing monitoring unit comprises a plurality of monitoring sensors and is used for collecting a plurality of element data reflecting the landslide state and transmitting the element data to the monitoring processor.
Optionally, the monitoring processor includes a low-power embedded microprocessor, and the low-power embedded microprocessor uses an STM 32L 071 chip.
Optionally, the multi-element sensing and monitoring unit includes a rainfall sensor, a displacement sensor, an inclination sensor, a soil water content sensor, an acceleration sensor, a soil pressure sensor, and a pore water pressure sensor.
Optionally, the rainfall sensor is in communication connection with the low-power-consumption embedded microprocessor through an IO port; the displacement sensor, the inclination angle sensor and the soil water content sensor are in communication connection with the low-power-consumption embedded microprocessor through an A/D conversion circuit; the acceleration sensor, the soil body pressure sensor and the pore water pressure sensor are in communication connection with the low-power-consumption embedded microprocessor through an RS485 circuit.
Optionally, the GNSS positioning apparatus includes one or more of beidou, GPS, G L ONASS.
Optionally, the three-dimensional space positioning monitoring unit further includes a signal analysis processor, configured to transfer a satellite signal received by the GNSS positioning apparatus to a radio frequency input end, amplify, filter, downconvert a signal in a radio frequency channel, and convert the signal into an intermediate frequency signal capable of performing analog conversion; the analog-to-digital converter is used for digitizing the intermediate frequency signal and then transmitting the digital signal to the baseband processor; the baseband processor is used for recovering the originally received positioning information and transmitting the positioning information to the monitoring processor.
Optionally, the landslide multi-element three-dimensional space monitoring system further comprises a wireless communication module, the monitoring processor is in communication connection with the monitoring cloud server through the wireless communication module, and the monitoring processor sends the acquired monitoring data to the monitoring cloud server.
Optionally, the wireless communication module is a 4G full network communication network, a beidou satellite communication network or an L oRa wireless ad hoc network.
The invention also provides a landslide multi-element three-dimensional space monitoring method, which is applied to the landslide multi-element three-dimensional space monitoring system and comprises the following steps:
s1, acquiring each monitoring data in the landslide multi-element three-dimensional space monitoring system, wherein the monitoring data comprises three-dimensional positioning information, elevation information and multiple element data reflecting landslide states;
s2, initializing weight: constructing a decision data training set T { (s { (S)1，y1)，(s2，y2)，…，(sN，yN) In which s isi(i ═ 1, 2, …, N) represents the respective monitoring data in the landslide multicomponent three-dimensional space monitoring system, yi(i-1, 2, …, N) represents the data processed by each monitoring sensorCharacteristic data, yiBelongs to a feature label set {1, 0 }; each training sample is initially given the same weight: 1/N, the initial weight set is as follows:
DM1＝(ω11，ω12，…，ω1i…，ω1N) (1)
wherein
S3, training the weak classifier: preliminarily setting M to be 1, 2 and …, wherein M is the iteration number, and acquiring the weak classifier F obtained in each iterationm(s) and calculating a corresponding classification error rate:
wherein the error ratemIs thatmQuilt Fm(s) the sum of the weights of the misclassified samples, calculating FmCoefficient of(s)
Wherein, αmRepresents Fm(s) the importance ratio in the final classifier, as can be seen from the formula,
αm≥0，αmWith followingmDecrease and increase; suppose that
The ensemble learning training is stopped; if the condition is satisfied, updating the weight distribution of the training set to perform the next iteration,
DMm+1＝(ωm+1，1，ωm+1，2，…，ωm+1，i…，ωm+1，N) (4)
wherein
Because of ZmIs a normalization factor such that DMm+1It becomes a probability distribution that,
s4, according to the weak classifier weight αmCombining weak classifiers, i.e.
Finally, the strong classifiers are formed,
according to the operation result of the strong classifier, the lower the fault tolerance rate of the strong classifier is, the higher the possibility of representing the occurrence of landslide and collapse disasters is.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: according to the landslide multi-element three-dimensional space monitoring system and method, a high-precision three-dimensional space positioning technology and a multi-element landslide real-time monitoring technology are organically integrated, so that space dimension information is added to monitoring data on the basis of original time dimension, and the effective information amount and the use value of the monitoring data are effectively improved; through a node deployment mode of a plurality of monitoring sensors, the point monitoring of an original monitoring mode can be expanded to the overall monitoring, a large amount of basic monitoring data can be obtained through a data fusion analysis technology, such as rainfall, soil water content, crack displacement, deep displacement, single-point surface deformation, stress and other landslide key elements and three-dimensional space coordinates of monitoring points, the overall three-dimensional space deformation condition of the landslide can also be obtained through analysis, and the overall control on the overall condition of the landslide body can be simply and efficiently realized by combining means such as the arrangement positions of the monitoring points; in addition, through the representation of the spatial position, the original monitoring parameters also have three-dimensional space vector attributes, and the method has higher value for application scenes of deep data mining and analysis, such as research on the transmission direction and speed of water, displacement, soil humidity, stress and the like.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used 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 inventive exercise.
FIG. 1 is a block diagram of a landslide multi-element three-dimensional space monitoring system according to an embodiment of the invention;
fig. 2 is a data processing flow chart of the landslide multi-element three-dimensional space monitoring system of 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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a landslide multi-element three-dimensional space monitoring system and a landslide multi-element three-dimensional space monitoring method, which organically integrate a three-dimensional space positioning technology and a multi-element real-time monitoring technology, comprehensively consider from the aspects of three-dimension and multi-dimension, realize all-dimensional, multi-element and high-precision monitoring on a landslide disaster body, and improve the reliability, stability and effectiveness of landslide disaster monitoring.
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.
Fig. 1 is a block diagram of a landslide multi-element three-dimensional space monitoring system according to an embodiment of the present invention, and as shown in fig. 1, the landslide multi-element three-dimensional space monitoring system according to the embodiment of the present invention includes: the system comprises a monitoring processor, a three-dimensional space positioning monitoring unit and a multi-element sensing monitoring unit, wherein the three-dimensional space positioning monitoring unit and the multi-element sensing monitoring unit are in communication connection with the monitoring processor, and the three-dimensional space positioning monitoring unit comprises a GNSS positioning device and is used for acquiring three-dimensional positioning information and elevation information of a landslide and transmitting the three-dimensional positioning information and the elevation information to the monitoring processor; the multi-element sensing monitoring unit comprises a plurality of monitoring sensors and is used for collecting a plurality of element data reflecting the landslide state and transmitting the element data to the monitoring processor.
The monitoring processor comprises a low-power-consumption embedded microprocessor, and the low-power-consumption embedded microprocessor adopts an STM 32L 071 chip.
The multi-element sensing and monitoring unit comprises a rainfall sensor, a displacement sensor, an inclination angle sensor, a soil water content sensor, an acceleration sensor, a soil body pressure sensor and a pore water pressure sensor; the rainfall sensor is in communication connection with the low-power-consumption embedded microprocessor through an IO port; the displacement sensor, the inclination angle sensor and the soil water content sensor are in communication connection with the low-power-consumption embedded microprocessor through an A/D conversion circuit; the acceleration sensor, the soil body pressure sensor and the pore water pressure sensor are in communication connection with the low-power-consumption embedded microprocessor through an RS485 circuit. And the RS485 circuit is used for connecting monitoring sensors far away from the system.
The GNSS positioning device comprises one or more of a Beidou, a GPS and a G L ONASS, the three-dimensional space positioning monitoring unit further comprises a signal analysis processor which is used for transmitting satellite signals received by the GNSS positioning device to a radio frequency input end, amplifying, filtering and down-converting signals in a radio frequency channel and converting the signals into intermediate frequency signals capable of analog conversion, an analog-to-digital converter which is used for digitizing the intermediate frequency signals and then transmitting the digital signals to a baseband processor, and the baseband processor is used for recovering originally received positioning information and transmitting the positioning information to the monitoring processor.
Based on basic data such as three-dimensional coordinates, speed and precise time service provided by a Global Navigation Satellite System (GNSS) such as Beidou, GPS, G L ONASS, a low-power-consumption and high-reliability satellite signal receiving device suitable for a geological disaster monitoring scene is researched and developed, functions such as receiving, filtering, storing and transmitting satellite signals are realized, a high-precision three-dimensional space positioning technology and related algorithms are researched, a three-dimensional space position resolving algorithm and a realizing mode based on the GNSS are researched, research on algorithms such as static RTK, dynamic RTK and multi-satellite data combined resolving is carried out, related software is compiled, algorithm performance and effect are continuously improved through means such as algorithm analysis, error propagation mechanism analysis and real scene test, millimeter-scale positioning precision is finally achieved, and four key technical problems are mainly solved for meeting requirements:
i: GNSS receiver data simulation and emulation
The GNSS data comprise pseudo-range information, carrier phase information, signal-to-noise ratio information, carrier phase ambiguity information and the like, the reconstruction of the pseudo-range information, the carrier phase information, the signal-to-noise ratio information, the carrier phase ambiguity information and the like needs high-precision GNSS ephemeris provided by the international GNSS service organization, the ephemeris recovery needs to use Lagrange's nine-order interpolation, and the clock correction needs to use second-order linear interpolation. The final product is a simulation file of observation data of the GNSS receiver with controllable parameters, and is used for checking the algorithm.
II: batch preprocessing of GNSS receiver measured data
The real GNSS observation data has a plurality of disturbance items, preprocessing is needed before use, low-quality information such as gross error, cycle slip and singular value is removed, and meanwhile, the preliminary weight of the observation value is obtained according to the prior analysis conclusion.
III: three-dimensional space positioning mechanism and scheme research
Establishing weights according to different observation angles, signal to noise ratios, ionosphere disturbance conditions and the like tracked to the GNSS satellite based on the preprocessed actual observation values, and performing centimeter-level high-precision single-point time-space reconstruction by using a weighted least square estimation theory; and aiming at the possible observation interruption, the state of the geological disaster area is predicted by means of extended Kalman filtering, and information is acquired to the maximum extent.
IV: high-precision space positioning calculation algorithm design and optimization
The method comprises the steps of calculating a baseline by means of a high-precision carrier phase observation value based on a known single-point coordinate, constructing a satellite-monitoring point double-difference model, calculating the integer ambiguity in the GNSS carrier phase observation value, wherein the current mainstream integer ambiguity calculation schemes comprise L AMBDA and W L/N L schemes, selecting and optimizing an optimal scheme aiming at a project, and finally obtaining a millimeter-scale baseline by means of a weighted least square estimation theory.
The landslide multi-element three-dimensional space monitoring system further comprises a wireless communication module, the monitoring processor is in communication connection with a monitoring cloud server through the wireless communication module, the monitoring processor sends acquired monitoring data to the monitoring cloud server, the wireless communication module is a 4G full-network communication network, a Beidou satellite communication network or an L oRa wireless ad-hoc network, if a mobile network is in the field, a communication circuit mainly sends the acquired monitoring data to the monitoring cloud server by means of a 4G full-network communication/NB-IoT/Beidou satellite, if the field is not provided with the mobile network, the on-site multi-element three-dimensional space monitoring system gathers the data to one node through L oRa self-networking, then data operation and calculation are carried out through the monitoring node, the monitoring data are transmitted to the monitoring cloud server through the Beidou satellite, in order to effectively solve the problems of unstable transmission, packet loss rate, no signal and the like of the existing monitoring data of disasters, the wireless transmission technology of the system is expanded, the wireless transmission technology of a new generation of 4G full-network communication, the geological transmission technology of L oRa and the like is suitable for being applied to the intelligent disaster monitoring cloud server under severe geological environment, the intelligent disaster monitoring system is constructed on the basis of cooperative work frequency, the intelligent disaster monitoring data acquisition and the working frequency, the intelligent disaster monitoring technology, the self-adaptive working efficiency judgment of the intelligent disaster monitoring system is realized, and the intelligent disaster monitoring system.
The low-power-consumption embedded microprocessor mainly realizes data acquisition, control and coordination management of disaster bodies. The data acquisition mainly comprises the steps of acquiring monitoring data (including a displacement sensor, an inclination angle sensor, a water content sensor, a pore water pressure sensor, a soil pressure sensor and the like) of a multi-element sensing monitoring unit and monitoring data (including specific three-dimensional positioning information and elevation information of disaster point positions) of a three-dimensional space positioning unit. The control is mainly to adjust the data sampling frequency and the uploading frequency of the multi-element sensing monitoring unit and the high-precision three-dimensional space positioning unit, realize the remote control and command interaction with the front-end monitoring sensor, and upload the collected multi-element sensing monitoring data and the high-precision three-dimensional space positioning information to the geological disaster monitoring and early warning cloud server. The coordination management unit mainly relates to system function parameter setting, which mainly comprises system working mode setting, power circuit management setting, system operation working frequency, data transmission working frequency range selection, remote transmission IP address, AD input port range (voltage and current) and the like.
Setting parameters of the low-power-consumption embedded microprocessor: the method is mainly used for setting the relevant working states of the multi-element three-dimensional space monitoring system, including data acquisition frequency, working modes, static/dynamic settlement modes and the like.
USB interface circuit of the embedded microprocessor of low-power consumption: the system is mainly used for programming a system program and keeping data connection with a PC (personal computer) end; the real-time clock circuit is mainly used for keeping the system on line in real time, calibrating time and ensuring the accuracy of sampling time, acquisition time and uploading time.
The landslide multi-element three-dimensional space monitoring system can be further provided with a power supply circuit, a wind-solar complementary power supply system is adopted, the monitoring system is charged in the field through wind energy and solar energy, the power supply of the landslide multi-element three-dimensional space monitoring system under a severe environment is effectively supplied, and the normal work of the multi-element three-dimensional space monitoring system is effectively guaranteed;
the landslide multi-element three-dimensional space monitoring system is also provided with a data storage circuit: the system is mainly used for storing three-dimensional space positioning information and multi-element monitoring data which are acquired by a multi-element three-dimensional space monitoring system in real time; the method mainly adopts a high-capacity TF memory card to store monitoring data; on the other hand, the EEPROM of the embedded low-power-consumption microprocessing can store the storage of the related set parameters of the system.
Fig. 2 shows a data processing process flow of the landslide multi-element three-dimensional space monitoring system of the present invention. In order to store the collected monitoring data into the database, when the collected monitoring data is stored into the database, the multi-element three-dimensional space monitoring data is subjected to data classification and respectively stored in a warehouse, because the two types of monitoring data have different processing modes, the data are divided into two types of data processing modes, and the general working flow is briefly described as follows;
after the system is powered on, hardware initialization is carried out on the monitoring system, then the multi-element three-dimensional space monitoring system starts to carry out data acquisition, whether an instruction is required to upload monitoring data or not is automatically inquired, CRC (cyclic redundancy check) code detection and calculation are carried out on the acquired monitoring data, if the data received at this time is judged to be in a correct format, the monitoring system is inquired whether to carry out warehousing operation on the monitoring data, and if the condition that data classification and identification of the multi-element three-dimensional space monitoring data are met; and identifying the character at the beginning of the data, if the character is GN, determining the data as three-dimensional space monitoring data, and performing data processing by the three-dimensional space monitoring data flow, and if the character at the beginning of the data is GP, performing data processing by the multi-element monitoring data flow. The processing flow is the data processing flow of the two columns at the right side in fig. 2. And then the two types of monitoring data are respectively stored in the database after being processed, and then the monitoring data are delivered to a final data analysis layer to judge the state of the disaster body so as to jointly judge whether the disaster body has a disaster or not.
The data processed and analyzed by the multi-element three-dimensional space monitoring system is in a multi-source heterogeneous data form, in order to cooperate with effective utilization and execution of each function of each data, a multi-source heterogeneous data fusion scheme is adopted, the system is divided into a database layer, a data fusion layer and a uniform application layer, the heterogeneous database layer is formed by heterogeneous databases, the data fusion layer is responsible for data access of heterogeneous data sources and coordination of information among the data sources, and the uniform application layer provides a visual data monitoring platform for users.
According to the landslide multi-element three-dimensional space monitoring system and method, a high-precision three-dimensional space positioning technology and a multi-element landslide real-time monitoring technology are organically integrated, so that space dimension information is added to monitoring data on the basis of original time dimension, and the effective information amount and the use value of the monitoring data are effectively improved; through a node deployment mode of a plurality of monitoring sensors, the point monitoring of an original monitoring mode can be expanded to the overall monitoring, a large amount of basic monitoring data can be obtained through a data fusion analysis technology, such as rainfall, soil water content, crack displacement, deep displacement, single-point surface deformation, stress and other landslide key elements and three-dimensional space coordinates of monitoring points, the overall three-dimensional space deformation condition of the landslide can also be obtained through analysis, and the overall control on the overall condition of the landslide body can be simply and efficiently realized by combining means such as the arrangement positions of the monitoring points; in addition, through the representation of the spatial position, the original monitoring parameters also have three-dimensional space vector attributes, and the method has higher value for application scenes of deep data mining and analysis, such as research on the transmission direction and speed of water, displacement, soil humidity, stress and the like.
The invention also provides a landslide multi-element three-dimensional space monitoring method, which is applied to the landslide multi-element three-dimensional space monitoring system and comprises the following steps:
s1, acquiring each monitoring data in the landslide multi-element three-dimensional space monitoring system, wherein the monitoring data comprises three-dimensional positioning information, elevation information and multiple element data reflecting landslide states;
s2, initializing weight: constructing a decision data training set T { (s { (S)1，y1)，(s2，y2)，…，(sN，yN) In which s isi(i ═ 1, 2, …, N) represents the respective monitoring data in the landslide multicomponent three-dimensional space monitoring system, yi(i-1, 2, …, N) represents characteristic data of each monitoring sensor after data processing, yiBelongs to a feature label set {1, 0 }; each training sample is initially given the same weight: 1/N, the weight set is as follows:
DM1＝(ω11，ω12，…，ω1i…，ω1N) (1)
wherein
The weight set represents that each monitoring data (a rainfall sensor, a displacement sensor, an inclination sensor, a soil water content sensor, an acceleration sensor, a soil body pressure sensor and a pore water pressure sensor) is endowed with the same weight at first, namely the proportion condition in the prediction and the forecast of the landslide disaster is carried out, the same weight is used at first, the correct proportion is higher and higher after iterative calculation, and the wrong proportion is less and less;
s3, training the weak classifier: preliminarily setting M to be 1, 2 and …, wherein M is the iteration number, 1 represents the first iteration, 2 represents the second iteration, and 3 represents the 3 rd iteration, and the weight value of each monitoring data set after each iteration is changed; obtaining weak classifiers F obtained by each iterationm(s) and calculating a corresponding classification error rate:
wherein the error ratemIs thatmQuilt Fm(s) the sum of the weights of the misclassified samples, calculating FmCoefficient of(s)
Wherein, αmRepresents Fm(s) the importance ratio in the final classifier, which means that the importance of the final landslide prediction is judged, for example, the ratio of the displacement sensor to the three-dimensional space positioning information monitoring data is heavy, and the importance of the displacement sensor to the three-dimensional space positioning information monitoring data is very high, so that the prediction result can be greatly influencedThe importance of the influence can indicate that the strong learning is correct;
as can be seen from the formula,
αm≥0，αmWith followingmDecrease and increase; suppose that
The ensemble learning training is stopped; if the condition is satisfied, updating the weight distribution of the training set to perform the next iteration,
DMm+1＝(ωm+1，1，ωm+1，2，…，ωm+1，i…，ωm+1，N) (4)
wherein
Because of ZmIs a normalization factor such that DMm+1It becomes a probability distribution that,
s4, according to the weak classifier weight αmCombining weak classifiers, i.e.
Finally, the strong classifiers are formed,
according to the operation result of the strong classifier, the lower the fault tolerance rate of the strong classifier is, the higher the possibility of representing the occurrence of landslide and collapse disasters is. The phenomenon that early warning accuracy becomes low due to the fact that error data of a single sensor appears can be effectively avoided through the strong classifier, and therefore decision early warning accuracy of landslide collapse disasters is effectively improved.
The landslide multi-element three-dimensional space monitoring method is characterized in that equivalence is given to N training samplesThe weight is 1/N. Then m iterations are performed, at training data (x)0，y0) Training a first weak learner, calculating a training error of each sample, and setting a threshold value, wherein when the training error of a sample is smaller than the threshold value, the training result of the sample is better, the weight of the sample is reduced, otherwise, the training result is poorer, and the weight of the sample is improved; in the second training, the samples (x) adjusted according to the weights1，y1) And training a second weak learner until m iterations are completed to obtain m weak learners. In order to obtain the final enhancement result from all weak learners, Adaboost assigns a weight to each weak learner, the weights are calculated based on the training error rate of each weak learner, and finally, the prediction result is obtained through the weighted integration of all weak learners.
The landslide disaster can be monitored in an all-around, multi-aspect and three-dimensional real-time online mode through the acquired multi-element three-dimensional space monitoring data, the integration and fusion of the multi-element monitoring data and the three-dimensional space positioning monitoring data are achieved through an effective data fusion algorithm, each monitoring sensor is different from each space positioning information data, a weak learning model of each monitoring sensor and each space positioning information is built, the weak learning model is enhanced through an integrated learning framework, and a prediction algorithm for the landslide multi-element three-dimensional space monitoring system data integration and fusion is achieved through an Adaboost integrated learning method based on machine learning.
The landslide multi-element three-dimensional space monitoring system and the method integrate and fuse a high-precision three-dimensional space positioning technology and a multi-element real-time monitoring technology from two levels of hardware and software to form uniform monitoring equipment with strong practicability, stability and reliability, and realize high-speed fusion and calculation of multi-element real-time monitoring data and three-dimensional space positioning data through embedded software control and bus level data intercommunication, so that dynamic motion data such as deformation speed, acceleration and the like are acquired in real time on the basis of basic monitoring data. By organically fusing a high-precision three-dimensional space positioning technology and a multi-factor real-time monitoring technology and uniformly considering the software and hardware circuit design of the two technologies, a corresponding monitoring module can be independently selected in actual monitoring to realize corresponding monitoring effect, the comprehensive, multi-aspect, three-dimensional and high-precision online monitoring on a landslide disaster body is really realized on the design thought, the instrument cost and the instrument power consumption of the landslide disaster body monitoring can be effectively reduced, in addition, three-dimensional space positioning information and multi-factor monitoring data information are organically fused on the software design, more accurate analysis and processing can be realized on the monitoring data on the whole, the monitoring data can realize cross complementation and intelligent linkage, the frequency of acquiring the data can be independently selected in a time-sharing, stage-grading and condition-grading manner, the utilization rate of the monitoring data is effectively improved, and more reliable data decision support is effectively provided for geological disaster experts, the method provides more reliable technical service for judging the landslide disaster body motion trend and monitoring and early warning, and is expected to play a larger and better role in the field of geological disaster monitoring and early warning.
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 (9)
1. A landslide multi-element three-dimensional space monitoring system, comprising: the system comprises a monitoring processor, a three-dimensional space positioning monitoring unit and a multi-element sensing monitoring unit, wherein the three-dimensional space positioning monitoring unit and the multi-element sensing monitoring unit are in communication connection with the monitoring processor, and the three-dimensional space positioning monitoring unit comprises a GNSS positioning device and is used for acquiring three-dimensional positioning information and elevation information of a landslide and transmitting the three-dimensional positioning information and the elevation information to the monitoring processor; the multi-element sensing monitoring unit comprises a plurality of monitoring sensors and is used for collecting a plurality of element data reflecting the landslide state and transmitting the element data to the monitoring processor.
2. The landslide multi-element three-dimensional space monitoring system according to claim 1 wherein said monitoring processor comprises a low power embedded microprocessor employing an STM 32L 071 chip.
3. The landslide multi-element three-dimensional space monitoring system according to claim 2 wherein the multi-element sensing monitoring unit comprises a rainfall sensor, a displacement sensor, an inclination sensor, a soil moisture content sensor, an acceleration sensor, a soil pressure sensor, a pore water pressure sensor.
4. The landslide multi-element three-dimensional space monitoring system according to claim 1, wherein the rainfall sensor is in communication connection with the low power consumption embedded microprocessor through an IO port; the displacement sensor, the inclination angle sensor and the soil water content sensor are in communication connection with the low-power-consumption embedded microprocessor through an A/D conversion circuit; the acceleration sensor, the soil body pressure sensor and the pore water pressure sensor are in communication connection with the low-power-consumption embedded microprocessor through an RS485 circuit.
5. The landslide multi-element three dimensional spatial monitoring system of claim 1 wherein said GNSS positioning means comprises one or more of beidou, GPS, G L ONASS.
6. The landslide multielement three-dimensional space monitoring system as recited in claim 1, wherein the three-dimensional space positioning monitoring unit further comprises a signal analysis processor for transmitting satellite signals received by the GNSS positioning device to a radio frequency input terminal, and amplifying, filtering and down-converting signals in a radio frequency channel into intermediate frequency signals capable of analog conversion; the analog-to-digital converter is used for digitizing the intermediate frequency signal and then transmitting the digital signal to the baseband processor; the baseband processor is used for recovering the originally received positioning information and transmitting the positioning information to the monitoring processor.
7. The landslide multi-element three-dimensional space monitoring system according to claim 1, further comprising a wireless communication module, wherein the monitoring processor is in communication connection with a monitoring cloud server through the wireless communication module, and the monitoring processor sends the obtained monitoring data to the monitoring cloud server.
8. The landslide multi-element three-dimensional space monitoring system according to claim 7, wherein the wireless communication module is a 4G full network communication network, a Beidou satellite communication network or an L oRa wireless ad hoc network.
9. A landslide multi-element three-dimensional space monitoring method applied to the landslide multi-element three-dimensional space monitoring system according to any one of claims 1-8, comprising the steps of:
s1, acquiring each monitoring data in the landslide multi-element three-dimensional space monitoring system, wherein the monitoring data comprises three-dimensional positioning information, elevation information and multiple element data reflecting landslide states;
s2, initializing weight: constructing a decision data training set T { (s { (S)1，y1)，(s2，y2)，…，(sN，yN) In which s isi(i ═ 1, 2, …, N) represents the respective monitoring data in the landslide multicomponent three-dimensional space monitoring system, yi(i-1, 2, …, N) represents characteristic data of each monitoring sensor after data processing, yiBelongs to a feature label set {1, 0 }; each training sample is initially given the same weight: 1/N, the initial weight set is as follows:
DM1＝(ω11，ω12，…，ω1i…，ω1N) (1)
wherein
S3, training the weak classifier: preliminarily setting M to 1, 2 and …, wherein M is the iteration numberObtaining the weak classifier F obtained from each iterationm(s) and calculating a corresponding classification error rate:
wherein the error ratemIs thatmQuilt Fm(s) the sum of the weights of the misclassified samples, calculating FmCoefficient of(s)
Wherein, αmRepresents Fm(s) the importance ratio in the final classifier, as can be seen from the formula,
αm≥0，αmWith followingmDecrease and increase; suppose that
The ensemble learning training is stopped; if the condition is satisfied, updating the weight distribution of the training set to perform the next iteration,
DMm+1＝(ωm+1，1，ωm+1，2，…，ωm+1，i…，ωm+1，N) (4)
wherein
Because of ZmIs a normalization factor such that DMm+1It becomes a probability distribution that,
s4, according to the weak classifier weight αmCombining weak classifiers, i.e.
Finally, the strong classifiers are formed,
according to the operation result of the strong classifier, the lower the fault tolerance rate of the strong classifier is, the higher the possibility of representing the occurrence of landslide and collapse disasters is.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010521577.2A CN111504392A (en) | 2020-06-10 | 2020-06-10 | Landslide multi-element three-dimensional space monitoring system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010521577.2A CN111504392A (en) | 2020-06-10 | 2020-06-10 | Landslide multi-element three-dimensional space monitoring system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111504392A true CN111504392A (en) | 2020-08-07 |
Family
ID=71878861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010521577.2A Pending CN111504392A (en) | 2020-06-10 | 2020-06-10 | Landslide multi-element three-dimensional space monitoring system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111504392A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112201011A (en) * | 2020-10-10 | 2021-01-08 | 上海华测导航技术股份有限公司 | Low-power consumption master control system with multiple sensor data acquisition and transmission |
| CN112258823A (en) * | 2020-10-21 | 2021-01-22 | 辽宁工程技术大学 | Strip mine landslide disaster remote monitoring system based on NB-IOT |
| CN112907904A (en) * | 2021-04-06 | 2021-06-04 | 成都万江港利科技股份有限公司 | Intelligent alarm control system and control method |
| CN113033108A (en) * | 2021-04-19 | 2021-06-25 | 昆明理工大学 | Side slope reliability judgment method based on AdaBoost algorithm |
| CN113053065A (en) * | 2021-03-31 | 2021-06-29 | 联通(江苏)产业互联网有限公司 | Expressway integrated monitoring and early warning platform system based on 5G and Beidou navigation system and use method thereof |
| CN113159431A (en) * | 2021-04-28 | 2021-07-23 | 宁夏回族自治区国土资源调查监测院 | Analysis early warning method and device based on ground disaster data |
| CN113625634A (en) * | 2021-08-12 | 2021-11-09 | 兰州大学 | Multi-physical-field landslide monitoring system and monitoring method |
| CN114049565A (en) * | 2021-11-08 | 2022-02-15 | 中国公路工程咨询集团有限公司 | Geological disaster identification method and device based on remote sensing image and DEM data |
| CN115343317A (en) * | 2022-08-15 | 2022-11-15 | 长安大学 | Loess landslide disaster comprehensive monitoring method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004191081A (en) * | 2002-12-09 | 2004-07-08 | Japan Radio Co Ltd | Satellite signal reception method and its reception apparatus |
| BRPI1001765A2 (en) * | 2010-02-26 | 2011-10-18 | Terranova Consultores Empresarial Associados Ltda | slope sliding monitoring and warning system |
| CN104021394A (en) * | 2014-06-05 | 2014-09-03 | 华北电力大学（保定） | Insulator image recognition method based on Adaboost algorithm |
| CN105844858A (en) * | 2016-04-05 | 2016-08-10 | 南信大影像技术工程（苏州）有限公司 | GNSS landslide monitoring and early warning system |
| CN105976569A (en) * | 2016-07-26 | 2016-09-28 | 长安大学 | Landslide hazard monitoring system and method |
| CN207423114U (en) * | 2017-11-30 | 2018-05-29 | 中国地质调查局水文地质环境地质调查中心 | A kind of landslide multi-point displacement intelligent monitoring device |
| CN110230990A (en) * | 2019-06-27 | 2019-09-13 | 北京讯腾智慧科技股份有限公司 | Gas ductwork geology deformation monitoring system and method |
-
2020
- 2020-06-10 CN CN202010521577.2A patent/CN111504392A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004191081A (en) * | 2002-12-09 | 2004-07-08 | Japan Radio Co Ltd | Satellite signal reception method and its reception apparatus |
| BRPI1001765A2 (en) * | 2010-02-26 | 2011-10-18 | Terranova Consultores Empresarial Associados Ltda | slope sliding monitoring and warning system |
| CN104021394A (en) * | 2014-06-05 | 2014-09-03 | 华北电力大学（保定） | Insulator image recognition method based on Adaboost algorithm |
| CN105844858A (en) * | 2016-04-05 | 2016-08-10 | 南信大影像技术工程（苏州）有限公司 | GNSS landslide monitoring and early warning system |
| CN105976569A (en) * | 2016-07-26 | 2016-09-28 | 长安大学 | Landslide hazard monitoring system and method |
| CN207423114U (en) * | 2017-11-30 | 2018-05-29 | 中国地质调查局水文地质环境地质调查中心 | A kind of landslide multi-point displacement intelligent monitoring device |
| CN110230990A (en) * | 2019-06-27 | 2019-09-13 | 北京讯腾智慧科技股份有限公司 | Gas ductwork geology deformation monitoring system and method |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112201011A (en) * | 2020-10-10 | 2021-01-08 | 上海华测导航技术股份有限公司 | Low-power consumption master control system with multiple sensor data acquisition and transmission |
| CN112258823A (en) * | 2020-10-21 | 2021-01-22 | 辽宁工程技术大学 | Strip mine landslide disaster remote monitoring system based on NB-IOT |
| CN113053065A (en) * | 2021-03-31 | 2021-06-29 | 联通(江苏)产业互联网有限公司 | Expressway integrated monitoring and early warning platform system based on 5G and Beidou navigation system and use method thereof |
| CN112907904A (en) * | 2021-04-06 | 2021-06-04 | 成都万江港利科技股份有限公司 | Intelligent alarm control system and control method |
| CN113033108A (en) * | 2021-04-19 | 2021-06-25 | 昆明理工大学 | Side slope reliability judgment method based on AdaBoost algorithm |
| CN113033108B (en) * | 2021-04-19 | 2022-05-27 | 昆明理工大学 | Side slope reliability judging method based on AdaBoost algorithm |
| CN113159431A (en) * | 2021-04-28 | 2021-07-23 | 宁夏回族自治区国土资源调查监测院 | Analysis early warning method and device based on ground disaster data |
| CN113625634A (en) * | 2021-08-12 | 2021-11-09 | 兰州大学 | Multi-physical-field landslide monitoring system and monitoring method |
| CN114049565A (en) * | 2021-11-08 | 2022-02-15 | 中国公路工程咨询集团有限公司 | Geological disaster identification method and device based on remote sensing image and DEM data |
| CN115343317A (en) * | 2022-08-15 | 2022-11-15 | 长安大学 | Loess landslide disaster comprehensive monitoring method and system |
| CN115343317B (en) * | 2022-08-15 | 2024-04-26 | 长安大学 | Loess landslide hazard comprehensive monitoring method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111504392A (en) | Landslide multi-element three-dimensional space monitoring system and method | |
| CN107703564B (en) | Rainfall prediction method and system and electronic equipment | |
| US11521022B2 (en) | Semantic state based sensor tracking and updating | |
| CN105588566A (en) | Indoor positioning system and method based on Bluetooth and MEMS (Micro-Electro-Mechanical Systems) fusion | |
| CN103728643B (en) | With the Big Dipper three network RTK blur level single epoch fixing means frequently that wide lane retrains | |
| CN112433236B (en) | Error model calibration method, device, equipment and computer readable storage medium | |
| CN109154664A (en) | The determination of navigation satellite track and clock with low delay clock correction value | |
| CN101110079A (en) | Digital globe antetype system | |
| CN109219762A (en) | The determination of the wide lane deviation of navigation satellite and over range adjust system and method | |
| CN102768361A (en) | GPS/INS combined positioning method based on genetic particle filtering and fuzzy neural network | |
| CN102902844A (en) | Sub-water basin partitioning method based on DEM (Dynamic Effect Model) data with large data quantity | |
| CN104574512A (en) | Multi-scale DEM (digital elevation model) construction method considering topographical semantic information | |
| CN111487657B (en) | Beidou real-time precise orbit determination method based on satellite perturbation | |
| CN108896962A (en) | Iteration localization method based on sound position fingerprint | |
| CN110991705B (en) | Urban extension prediction method and system based on deep learning | |
| CN117648863B (en) | Regional water supply capacity acquisition method under special drought conditions based on water system communication | |
| CN202770447U (en) | Indoor intelligent positioning navigation system | |
| CN116090596A (en) | Typhoon intensity prediction method, typhoon intensity prediction device, typhoon intensity prediction equipment and storage medium | |
| CN113543026B (en) | Multi-floor indoor positioning method based on radial basis function network | |
| CN117516517A (en) | Passive fusion positioning method and system in indoor environment and electronic equipment | |
| CN111854745A (en) | Clock prediction method based on Internet of things indoor positioning | |
| CN109633531A (en) | Wireless sensor network node positioning system under composite noise condition | |
| CN115829163A (en) | Multi-mode integration-based runoff prediction method and system for middle and lower reaches of Yangtze river | |
| Zhang et al. | Underground mining intelligent response and rescue systems | |
| Cui et al. | Design and development of GPS navigation and positioning system based on AutoCAD platform |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200921 Address after: 300300 5-1-101, Warner Jinghu garden, Dongli District, Tianjin Applicant after: Hydrogeological and Environmental Geological Survey Center of China Geological Survey Address before: 071000 No. 71 middle 1305 Road, Baoding, Hebei Applicant before: Hydrogeological and Environmental Geological Survey Center of China Geological Survey Applicant before: UNIVERSITY OF ELECTRONIC SCIENCE AND TECHNOLOGY |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200807 |