CN116861317A - Cell waterlogging early warning method and system based on BP neural network - Google Patents
Cell waterlogging early warning method and system based on BP neural network Download PDFInfo
- Publication number
- CN116861317A CN116861317A CN202311129638.0A CN202311129638A CN116861317A CN 116861317 A CN116861317 A CN 116861317A CN 202311129638 A CN202311129638 A CN 202311129638A CN 116861317 A CN116861317 A CN 116861317A
- Authority
- CN
- China
- Prior art keywords
- waterlogging
- neural network
- cell
- early warning
- data
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 67
- 238000003062 neural network model Methods 0.000 claims abstract description 55
- 238000012544 monitoring process Methods 0.000 claims abstract description 20
- 230000002265 prevention Effects 0.000 claims abstract description 11
- 238000007781 pre-processing Methods 0.000 claims abstract description 9
- 238000012360 testing method Methods 0.000 claims abstract description 7
- 210000004027 cell Anatomy 0.000 claims description 95
- 210000002569 neuron Anatomy 0.000 claims description 42
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 18
- 238000012795 verification Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 5
- 239000000463 material Substances 0.000 claims description 4
- 230000001537 neural effect Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims description 3
- 230000001960 triggered effect Effects 0.000 claims description 3
- 238000005086 pumping Methods 0.000 claims description 2
- 230000010485 coping Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000010223 real-time analysis Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- E—FIXED CONSTRUCTIONS
- E03—WATER SUPPLY; SEWERAGE
- E03F—SEWERS; CESSPOOLS
- E03F1/00—Methods, systems, or installations for draining-off sewage or storm water
-
- E—FIXED CONSTRUCTIONS
- E03—WATER SUPPLY; SEWERAGE
- E03F—SEWERS; CESSPOOLS
- E03F7/00—Other installations or implements for operating sewer systems, e.g. for preventing or indicating stoppage; Emptying cesspools
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/10—Alarms for ensuring the safety of persons responsive to calamitous events, e.g. tornados or earthquakes
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A10/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE at coastal zones; at river basins
- Y02A10/40—Controlling or monitoring, e.g. of flood or hurricane; Forecasting, e.g. risk assessment or mapping
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Hydrology & Water Resources (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Software Systems (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Emergency Management (AREA)
- Water Supply & Treatment (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- Biomedical Technology (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Environmental & Geological Engineering (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geology (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a cell waterlogging early warning method and system based on a BP neural network, comprising the following steps: obtaining a plurality of groups of sample data by using cell historical waterlogging data, taking meteorological information data, geographic information data and drainage information data as input parameters, taking the occurrence probability of waterlogging and the influence range of waterlogging as output parameters, preprocessing, determining structural parameters of a BP neural network, constructing a BP neural network model, training, adjusting the parameters of the BP neural network model, testing, obtaining a trained BP neural network model, predicting cell waterlogging, obtaining the occurrence probability of waterlogging and the output parameters of the influence range of waterlogging, generating waterlogging early warning information of different grades, and starting corresponding flood control and waterlogging prevention measures; the invention relates to the technical field of hydrological monitoring and early warning monitoring, which can timely and accurately predict the occurrence probability and influence range of residential flooding and issue early warning, so as to improve the coping capacity of residential community residents and reduce the influence of the flooding on the living and infrastructure of the residential community residents.
Description
Technical Field
The invention relates to the technical field of hydrological monitoring and early warning monitoring, in particular to a cell waterlogging early warning method and system based on a BP neural network.
Background
With the rapid development of social economy and the continuous acceleration of the urban process, the problem of urban waterlogging is increasingly prominent, and the waterlogging has great influence on urban infrastructure, mass life and ecological environment; the cell waterlogging is one of the common problems in the urban process, and has great influence on urban traffic, cell houses, cell resident life and the like.
However, the problems of the current waterlogging prevention means are mainly manifested in three aspects: firstly, the current early warning information is issued mainly depending on weather station rainfall statistics for prediction, but the distribution of the rainfall stations is set according to weather detection requirements and is not overlapped with the waterlogging points of each district in the city, so that the source of the early warning information obtained by a flood prevention emergency department is not accurate enough, and the accurate early warning of the waterlogging points of the district cannot be carried out; secondly, when the rainfall reaches a certain level, the meteorological department issues an early warning to inform the urban flood control management department, the flood control management department takes emergency measures again, and the district flood control response is passive; finally, the weather department usually uses the modes of calling and sending information when issuing and transmitting the early warning information, the efficiency is low, the time efficiency is poor, the real-time early warning can not be carried out on the district, and the relevant department starts a large number of staff to attend to the on-site monitoring investigation whenever waterlogging occurs, so that a large amount of manpower and material resources are consumed.
With the continuous development of computer technology, waterlogging early warning systems based on artificial intelligence gradually become research hotspots. The waterlogging early warning system based on the BP neural network has the advantages of high precision, high efficiency and the like, and becomes one of the key points of research.
Therefore, how to study and design a method for accurately early warning the cell waterlogging in real time through a BP neural network so as to take different flood control and waterlogging prevention measures according to different grades of early warning information, and improving the capacity of the city to cope with the waterlogging is a problem which needs to be solved by the technicians in the field.
Disclosure of Invention
In view of the above, the invention provides a cell waterlogging early warning method and a cell waterlogging early warning system based on a BP neural network, which solve the problem that a cell can accurately and timely early warn a sudden rainfall and start corresponding emergency measures for rainfall of different degrees.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a cell waterlogging early warning method based on BP neural network comprises the following steps:
s1, acquiring a plurality of groups of sample data by using cell historical waterlogging data, wherein the waterlogging data comprises meteorological information data, geographic information data and drainage information data.
S2, determining structural parameters of the BP neural network: taking meteorological information data, geographic information data and drainage information data as input parameters, taking waterlogging occurrence probability and waterlogging influence range as output parameters, and preprocessing;
s3, constructing a BP neural network model, dividing the preprocessed multiple groups of data into a training group and a verification group, training the BP neural network model by using cell history waterlogging data in the training group, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in the verification group after training is completed to obtain a trained BP neural network model;
s4, acquiring real-time meteorological information data of the cell, predicting the inland inundation of the cell by using a trained BP neural network model, obtaining output parameters of incidence probability and influence range of the inland inundation, judging whether inland inundation is generated or not and whether the inland inundation range is generated, and generating inland early warning information of different grades;
s5, starting corresponding flood control measures according to the flood warning information of different grades.
Preferably, the weather information data includes rainfall, rainfall intensity, and rainfall duration; the geographic information data comprises cell elevation and cell location; the drainage information data comprise drainage port distribution, pipe network layout and drainage pump stations.
Preferably, the preprocessing is to normalize various data of rainfall, rainfall intensity, rainfall duration, cell elevation, cell position, drainage distribution, pipe network layout and drainage pumping station before training and learning:
wherein ,values after normalization for various data, x i As the original value +.>Is the minimum value in various data +.>Is the maximum value in various data.
Preferably, the constructed BP neural network model comprises an input layer, an implicit layer and an output layer;
the number of the input layer nodes is 8, the number of the intermediate hidden layer nodes is 10, the number of the output layer nodes is 2, the linear relu function is an activation function of the intermediate hidden layer neurons, and the S-type sigmoid function is an output layer neural function.
Preferably, the specific content of training is as follows:
the input vector X= { rainfall, rainfall intensity, rainfall duration, cell elevation, cell position, drainage distribution, pipe network layout and drainage pump station }, through weights and thresholds among layers, forward transmission is carried out until the output layer, the output result Y= { waterlogging occurrence probability and waterlogging influence range }, an error value between an expected output value and an actual output value is calculated, if the error value does not meet a preset convergence value, a reverse feedback process is carried out, the weights and thresholds among the layers are modified, next training is carried out, and training is stopped until the preset convergence value is met.
Preferably, the intermediate hidden layer output in the training process is:
wherein ,an ith output that is an s-th layer neuron; n is n s-1 Is the number of the neurons of the s-1 layer; />A link weight between the jth neuron of the s-1 layer and the ith neuron of the s layer; />Bias for the s-layer ith neuron, < ->Is the activation function of neurons, siThe gmoid function is:
output layer output variable:
wherein ,output of the kth neuron as an output layer; />The number of neurons for the s-th hidden layer; the output of the ith neuron which is the s-th hidden layer; />The weight between the ith neuron of the s hidden layer and the kth neuron of the output layer is given; />A threshold value for the kth neuron of the output layer;
the error between the actual output value and the expected output value of the sample is:
wherein t is the training times; m is the number of output vectors of the output layer; y is k Output of the kth neuron as an output layer;is the kth desired output;
adjusting the weight and threshold of the network according to the normal training principle, and performing feedforward training again to obtainIs provided with->For the difference between the error of the training output value and the expected value of the two times before and after, if +.>Beta is the training precision of the neural network, the weight and the threshold value are updated, if +.>The training is ended.
Preferably, the waterlogging early warning information comprises an early warning level, an early warning area and early warning time.
Preferably, the early warning level is divided into 3 early warning levels of I level early warning, II level early warning and III level early warning, and the specific content of corresponding flood control and waterlogging prevention measures comprises:
III, early warning: when judging that low-intensity rainfall exists but no waterlogging point exists through the BP neural network model, sending rainfall alarm information to residents in a community to remind the residents of paying attention to safety, closing doors and windows, and paying attention to precautions;
II, early warning: when the BP neural network model judges that medium-intensity rainfall exists and waterlogging points are generated, cell monitoring and water level monitoring alarm are started, water level change conditions are obtained and analyzed in real time, when the water level exceeds a first preset threshold value, an alarm signal is automatically triggered, a cell unit and a flood control baffle flood control facility of an underground garage are started to rapidly lift, and when the water level exceeds a second preset threshold value, a cell drainage pump is started to drain water outside a cell;
i-stage early warning: when the BP neural network model judges that high-intensity rainfall exists and large-area waterlogging points are generated, the I-level early warning and the II-level early warning are started simultaneously, people are evacuated in the waterlogging points judged through BP neural network simulation, power supplies in the cells are cut off to prevent electric shock accidents, and meanwhile emergency rescue departments are notified of the emergency and request for materials and manpower support.
Preferably, the district monitoring is to install the surveillance camera head in the area of the easy ponding of outlet, low-lying area and underground garage, obtain the water level change condition in real time through the surveillance camera head, and the water level monitoring is reported to the police for carrying out real-time analysis to the water level information of surveillance camera head transmission.
The cell waterlogging early warning system based on the BP neural network comprises a data acquisition module, a data processing and analyzing module, a model construction and training module and an early warning information release module;
the data acquisition module is used for acquiring historical waterlogging data of the cell to obtain a plurality of groups of sample data, wherein the waterlogging data comprises meteorological information data, geographic information data and drainage information data, and is also used for acquiring real-time meteorological information data of the cell.
The data processing and analyzing module is used for determining structural parameters of the BP neural network according to the plurality of groups of sample data: taking meteorological information data, geographic information data and drainage information data as input parameters, taking waterlogging occurrence probability and waterlogging influence range as output parameters, and preprocessing;
the model construction and training module is used for constructing a BP neural network model, dividing the preprocessed multiple groups of data into a training group and a verification group, training the BP neural network model by using cell history waterlogging data in the training group, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in the verification group after training is completed to obtain a trained BP neural network model;
the trained BP neural network model is used for predicting the cell waterlogging according to the real-time meteorological information data of the cell, and obtaining the occurrence probability of the waterlogging and the output parameters of the waterlogging influence range;
and the early warning information release module judges whether the waterlogging and the waterlogging range can be generated according to the output parameters, obtains the waterlogging early warning information of different grades, and starts corresponding flood control and waterlogging prevention measures according to the waterlogging early warning information of different grades.
Compared with the prior art, the invention discloses a cell waterlogging early warning method and system based on BP neural network, which has the following advantages:
(1) Comprehensive utilization of multi-source data: by collecting real-time weather information, geographic information data and drainage information data in the cell, comprehensive assessment of waterlogging risk is realized, and accuracy and instantaneity of early warning of the waterlogging in the cell are improved by comprehensively utilizing multi-source data;
(2) Efficient data processing and analysis: adopting advanced data processing and analysis technology to predict rainfall capacity, evaluate cell drainage capacity, evaluate risk of waterlogging and the like, and providing scientific basis for building a waterlogging early warning model;
(3) And (3) building an accurate early warning model: by means of the method of calculating the occurrence probability of the waterlogging, predicting the influence range of the waterlogging and classifying the early warning grades, an accurate waterlogging early warning model is constructed, the occurrence probability and the influence range of the waterlogging are accurately predicted, and early warning information of corresponding grades is generated;
(4) Real-time early warning information release: judging whether waterlogging occurs or not according to the result of the early warning model construction module, generating waterlogging early warning information in real time, and adopting different flood control and waterlogging prevention measures according to the early warning information of different grades, so that the life and property safety of residents is practically protected.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of a cell waterlogging early warning method based on a BP neural network;
FIG. 2 is a schematic diagram of a BP neural network prediction model provided by the invention;
FIG. 3 is a flow chart of training neural network models provided by the present invention;
FIG. 4 is a schematic diagram of prediction of occurrence probability of waterlogging according to the present invention;
FIG. 5 is a schematic diagram of prediction of the waterlogging influence range provided by the invention;
fig. 6 is a schematic diagram of a cell waterlogging early warning system based on a BP neural network.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The embodiment of the invention discloses a cell waterlogging early warning method based on a BP neural network, as shown in figure 1, comprising the following steps of:
s1, acquiring a plurality of groups of sample data by using cell historical waterlogging data, wherein the waterlogging data comprises meteorological information data, geographic information data and drainage information data.
S2, determining structural parameters of the BP neural network: taking meteorological information data, geographic information data and drainage information data as input parameters, taking waterlogging occurrence probability and waterlogging influence range as output parameters, and preprocessing;
s3, constructing a BP neural network model, dividing the preprocessed multiple groups of data into a training group and a verification group, training the BP neural network model by using cell history waterlogging data in the training group, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in the verification group after training is completed to obtain a trained BP neural network prediction model;
s4, acquiring real-time meteorological information data of the cell, predicting the inland inundation of the cell by using a trained BP neural network model, obtaining output parameters of incidence probability and influence range of the inland inundation, judging whether inland inundation is generated or not and whether the inland inundation range is generated, and generating inland early warning information of different grades;
s5, starting corresponding flood control measures according to the flood warning information of different grades.
In order to further implement the above technical solution, the weather information data includes rainfall, rainfall intensity and rainfall duration; the geographic information data comprises cell elevation and cell location; the drainage information data comprise drainage port distribution, pipe network layout and drainage pump stations.
The rainfall data of the last ten years of the area of a certain district comprises rainfall, rainfall intensity and rainfall duration; secondly, collecting the landform, the underlying surface, the drainage pump station and the position of the drainage pipe network of the district; and then confirming the power of the district drainage pump station and the pipeline caliber of the drainage pipe network to judge the district drainage capacity.
In order to further implement the technical scheme, the pretreatment is to normalize rainfall, rainfall intensity, rainfall duration, cell elevation, cell position, discharge distribution, pipe network layout and various data of a drainage pump station before training and learning:
wherein ,values after normalization for various data, x i As the original value +.>Is the minimum value in various data +.>Is the maximum value in various data.
In this embodiment, the normalized sets of data include 8 input data and 2 output data, and the training set and the verification set are divided by 10-fold cross verification, where the ratio of the two sets is 8:2, repeating 10 times to average.
Because the waterlogging ponding depth is related to rainfall conditions, cell geographical positions and drainage capacity of cells, and belongs to the nonlinear regression problem of multiple influencing factors, a feedback neural network model, namely a BP neural network model is adopted to train and predict the waterlogging condition.
In order to further implement the above technical solution, as shown in fig. 2, the constructed BP neural network model includes an input layer, an hidden layer and an output layer;
the number of the input layer nodes is 8, the number of the intermediate hidden layer nodes is 10, the number of the output layer nodes is 2, the linear relu function is an activation function of the intermediate hidden layer neurons, and the S-type sigmoid function is an output layer neural function.
In order to further implement the above technical solution, as shown in fig. 3, the specific content of training is:
the input vector X= { rainfall, rainfall intensity, rainfall duration, cell elevation, cell position, drainage distribution, pipe network layout and drainage pump station }, through weights and thresholds among layers, forward transmission is carried out until the output layer, the output result Y= { waterlogging occurrence probability and waterlogging influence range }, an error value between an expected output value and an actual output value is calculated, if the error value does not meet a preset convergence value, a reverse feedback process is carried out, the weights and thresholds among the layers are modified, next training is carried out, and training is stopped until the preset convergence value is met.
In order to further implement the above technical solution, the output of the intermediate hidden layer in the training process is:
wherein ,an ith output that is an s-th layer neuron; n is n s-1 Is the number of the neurons of the s-1 layer; />A link weight between the jth neuron of the s-1 layer and the ith neuron of the s layer; />Bias for the s-layer ith neuron, < ->Is the activation function of the neuron, which is the sigmoid function:
output layer output variable:
wherein ,output of the kth neuron as an output layer; />The number of neurons for the s-th hidden layer; the output of the ith neuron which is the s-th hidden layer; />The weight between the ith neuron of the s hidden layer and the kth neuron of the output layer is given; />A threshold value for the kth neuron of the output layer;
the error between the actual output value and the expected output value of the sample is:
wherein t is the training times; m is the number of output vectors of the output layer; y is k Output of the kth neuron as an output layer;is the kth desired output;
adjusting the weight and threshold of the network according to the normal training principle, and performing feedforward training again to obtainIs provided with->For the difference between the error of the training output value and the expected value of the two times before and after, if +.>Beta is the training precision of the neural network, the weight and the threshold value are updated, if +.>The training is ended.
In this embodiment, after the network model training is completed, the verification group data is brought into, the accuracy of the neural network model is verified, if the expected error verified by the cell waterlogging prediction model is within 5%, the neural network model can well approach the actual situation, and the prediction parameters of the waterlogging occurrence probability and the waterlogging influence range are output:
as shown in fig. 4, the probability fitting error of the BP neural network successfully trained to the waterlogging is 0.014458, which is smaller than 0.05, and the network model realizes the prediction of whether to generate the waterlogging;
as shown in fig. 5, the fitting error of the BP neural network successfully trained to the waterlogging influence range is 0.0023689, which is smaller than 0.05, and the network model realizes the prediction of the waterlogging range.
In order to further implement the technical scheme, the waterlogging early warning information comprises an early warning level, an early warning area and early warning time.
In order to further implement the technical scheme, the early warning level is divided into 3 early warning levels of I level early warning, II level early warning and III level early warning, and the specific content of the corresponding flood control and waterlogging prevention measures comprises:
III, early warning: when the BP neural network model judges that low-intensity rainfall exists but no waterlogging point exists, rainfall alarm information is sent to residents in a community in a mode of short messages, mobile phone application and the like, and the residents are reminded of paying attention to safety, closing doors and windows and paying attention to precautions;
II, early warning: when the BP neural network model judges that medium-intensity rainfall exists and waterlogging points are generated, cell monitoring and water level monitoring alarm are started, water level change conditions are obtained and analyzed in real time, when the water level exceeds a first preset threshold value by 30cm, alarm signals are automatically triggered, a cell unit and a flood control baffle flood control facility of a underground garage are started to rapidly lift, when the water level exceeds a second preset threshold value by 50cm, a cell drainage pump is started, and drainage is carried out outside the cell;
i-stage early warning: when the BP neural network model judges that high-intensity rainfall exists and large-area waterlogging points are generated, the I-level early warning and the II-level early warning are started simultaneously, people are evacuated in the waterlogging points judged through BP neural network simulation, power supplies in the cells are cut off to prevent electric shock accidents, and meanwhile emergency rescue departments are notified of the emergency and request for materials and manpower support.
In order to further implement the technical scheme, the district monitoring is to install the surveillance camera head in the area of the easy ponding area of outlet, low-lying area and underground garage, obtains the water level change condition in real time through the surveillance camera head, and the water level monitoring is reported to the police for carrying out real-time analysis to the water level information of surveillance camera head transmission.
The cell waterlogging early warning system based on the BP neural network, as shown in figure 6, is based on a cell waterlogging early warning method based on the BP neural network, and comprises a data acquisition module, a data processing and analyzing module, a model construction and training module and an early warning information release module;
the data acquisition module is used for acquiring historical waterlogging data of the cell to obtain a plurality of groups of sample data, wherein the waterlogging data comprises meteorological information data, geographic information data and drainage information data, and is also used for acquiring real-time meteorological information data of the cell.
The data processing and analyzing module is used for determining structural parameters of the BP neural network according to the plurality of groups of sample data: taking meteorological information data, geographic information data and drainage information data as input parameters, taking waterlogging occurrence probability and waterlogging influence range as output parameters, and preprocessing;
the model construction and training module is used for constructing a BP neural network model, dividing the preprocessed multiple groups of data into a training group and a verification group, training the BP neural network model by using cell history waterlogging data in the training group, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in the verification group after training is completed to obtain a trained BP neural network model;
the trained BP neural network model is used for predicting the cell waterlogging according to the real-time meteorological information data of the cell, and obtaining the occurrence probability of the waterlogging and the output parameters of the waterlogging influence range;
and the early warning information release module judges whether the waterlogging and the waterlogging range can be generated according to the output parameters, obtains the waterlogging early warning information of different grades, and starts corresponding flood control and waterlogging prevention measures according to the waterlogging early warning information of different grades.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. The cell waterlogging early warning method based on the BP neural network is characterized by comprising the following steps of:
s1, acquiring a plurality of groups of sample data by using cell historical waterlogging data, wherein the waterlogging data comprises meteorological information data, geographic information data and drainage information data;
s2, determining structural parameters of the BP neural network: taking meteorological information data, geographic information data and drainage information data as input parameters, taking waterlogging occurrence probability and waterlogging influence range as output parameters, and preprocessing;
s3, constructing a BP neural network model, dividing the preprocessed multiple groups of data into a training group and a verification group, training the BP neural network model by using cell history waterlogging data in the training group, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in the verification group after training is completed to obtain a trained BP neural network model;
s4, acquiring real-time meteorological information data of the cell, predicting the inland inundation of the cell by using a trained BP neural network model, obtaining output parameters of incidence probability and influence range of the inland inundation, judging whether inland inundation is generated or not and whether the inland inundation range is generated, and generating inland early warning information of different grades;
s5, starting corresponding flood control measures according to the flood warning information of different grades.
2. The cell waterlogging early warning method based on the BP neural network according to claim 1, wherein the meteorological information data comprise rainfall, rainfall intensity and rainfall duration; the geographic information data comprises cell elevation and cell location; the drainage information data comprise drainage port distribution, pipe network layout and drainage pump stations.
3. The cell waterlogging early warning method based on the BP neural network according to claim 2, wherein the preprocessing is to normalize various data of rainfall, rainfall intensity, rainfall duration, cell elevation, cell position, drainage distribution, pipe network layout and drainage pumping station before training and learning:
；
wherein ,values after normalization for various data, x i As the original value +.>Is the minimum value in various types of data,is the maximum value in various data.
4. The cell waterlogging early warning method based on the BP neural network according to claim 1, wherein the constructed BP neural network model comprises an input layer, an implicit layer and an output layer;
the number of the input layer nodes is 8, the number of the intermediate hidden layer nodes is 10, the number of the output layer nodes is 2, the linear relu function is an activation function of the intermediate hidden layer neurons, and the S-type sigmoid function is an output layer neural function.
5. The cell waterlogging early warning method based on the BP neural network according to claim 4, wherein the training comprises the following specific contents:
the input vector X= { rainfall, rainfall intensity, rainfall duration, cell elevation, cell position, drainage distribution, pipe network layout and drainage pump station }, through weights and thresholds among layers, forward transmission is carried out until the output layer, the output result Y= { waterlogging occurrence probability and waterlogging influence range }, an error value between an expected output value and an actual output value is calculated, if the error value does not meet a preset convergence value, a reverse feedback process is carried out, the weights and thresholds among the layers are modified, next training is carried out, and training is stopped until the preset convergence value is met.
6. The cell waterlogging early warning method based on the BP neural network according to claim 5, wherein the output of an intermediate hidden layer in the training process is as follows:
；
wherein ,an ith output that is an s-th layer neuron; n is n s-1 Is the number of the neurons of the s-1 layer; />A link weight between the jth neuron of the s-1 layer and the ith neuron of the s layer; />Bias for the s-layer ith neuron, < ->Is the activation function of the neuron, which is the sigmoid function:
；
output layer output variable:
；
wherein ,output of the kth neuron as an output layer; />The number of neurons for the s-th hidden layer; the output of the ith neuron which is the s-th hidden layer; />The weight between the ith neuron of the s hidden layer and the kth neuron of the output layer is given;a threshold value for the kth neuron of the output layer;
the error between the actual output value and the expected output value of the sample is:
；
wherein t is the training times; m is the number of output vectors of the output layer; y is k Output of the kth neuron as an output layer;is the kth desired output;
adjusting the weight and threshold of the network according to the normal training principle, and performing feedforward training again to obtainIs provided withFor the difference between the error of the training output value and the expected value of the two times before and after, if +.>Beta is the training precision of the neural network, the weight and the threshold value are updated, if +.>The training is ended.
7. The cell waterlogging early-warning method based on the BP neural network according to claim 1, wherein the waterlogging early-warning information comprises an early-warning level, an early-warning area and early-warning time.
8. The cell waterlogging early warning method based on the BP neural network according to claim 1, wherein the early warning levels are divided into 3 early warning levels of level I early warning, level II early warning and level III early warning, and the corresponding flood control and waterlogging prevention measures specifically comprise:
III, early warning: when judging that low-intensity rainfall exists but no waterlogging point exists through the BP neural network model, sending rainfall alarm information to residents in a community to remind the residents of paying attention to safety, closing doors and windows, and paying attention to precautions;
II, early warning: when the BP neural network model judges that medium-intensity rainfall exists and waterlogging points are generated, cell monitoring and water level monitoring alarm are started, water level change conditions are obtained and analyzed in real time, when the water level exceeds a first preset threshold value, an alarm signal is automatically triggered, a cell unit and a flood control baffle flood control facility of an underground garage are started to rapidly lift, and when the water level exceeds a second preset threshold value, a cell drainage pump is started to drain water outside a cell;
i-stage early warning: when the BP neural network model judges that high-intensity rainfall exists and large-area waterlogging points are generated, the I-level early warning and the II-level early warning are started simultaneously, people are evacuated in the waterlogging points judged through BP neural network simulation, power supplies in the cells are cut off to prevent electric shock accidents, and meanwhile emergency rescue departments are notified of the emergency and request for materials and manpower support.
9. The cell waterlogging early warning method based on the BP neural network, which is characterized in that cell monitoring is to install monitoring cameras at places of water outlets, low-lying zones and easily-accumulated areas of underground garages, water level change conditions are obtained in real time through the monitoring cameras, and water level monitoring alarm is to analyze water level information transmitted by the monitoring cameras in real time.
10. The cell waterlogging early warning system based on the BP neural network is characterized by comprising a data acquisition module, a data processing and analyzing module, a model construction and training module and an early warning information release module, wherein the cell waterlogging early warning method based on the BP neural network is based on any one of claims 1-9;
the data acquisition module is used for acquiring historical waterlogging data of the cell to obtain a plurality of groups of sample data, wherein the waterlogging data comprises meteorological information data, geographic information data and drainage information data, and is also used for acquiring real-time meteorological information data of the cell;
the data processing and analyzing module is used for determining structural parameters of the BP neural network according to the plurality of groups of sample data: taking meteorological information data, geographic information data and drainage information data as input parameters, taking waterlogging occurrence probability and waterlogging influence range as output parameters, and preprocessing;
the model construction and training module is used for constructing a BP neural network model, dividing the preprocessed multiple groups of data into a training group and a verification group, training the BP neural network model by using cell history waterlogging data in the training group, adjusting network parameters of the BP neural network model, and testing the trained BP neural network model by using data in the verification group after training is completed to obtain a trained BP neural network model;
the trained BP neural network model is used for predicting the cell waterlogging according to the real-time meteorological information data of the cell, and obtaining the occurrence probability of the waterlogging and the output parameters of the waterlogging influence range;
and the early warning information release module judges whether the waterlogging and the waterlogging range can be generated according to the output parameters, obtains the waterlogging early warning information of different grades, and starts corresponding flood control and waterlogging prevention measures according to the waterlogging early warning information of different grades.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311129638.0A CN116861317A (en) | 2023-09-04 | 2023-09-04 | Cell waterlogging early warning method and system based on BP neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311129638.0A CN116861317A (en) | 2023-09-04 | 2023-09-04 | Cell waterlogging early warning method and system based on BP neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116861317A true CN116861317A (en) | 2023-10-10 |
Family
ID=88221976
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311129638.0A Pending CN116861317A (en) | 2023-09-04 | 2023-09-04 | Cell waterlogging early warning method and system based on BP neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116861317A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222427A (en) * | 2019-06-10 | 2019-09-10 | 河海大学 | A kind of analysis method based on mathematical model urban waterlogging |
| CN111582541A (en) * | 2020-03-27 | 2020-08-25 | 合肥学院 | Firefly algorithm-based inland inundation model prediction method |
| CN113344305A (en) * | 2021-08-05 | 2021-09-03 | 中国水利水电科学研究院 | Rapid prediction method for rainstorm waterlogging event |
| CN114048943A (en) * | 2021-10-08 | 2022-02-15 | 广州珠矶科技有限公司 | Urban waterlogging analysis method, device and medium based on unmanned aerial vehicle BIM and SWMM |
| CN114254561A (en) * | 2021-12-17 | 2022-03-29 | 奥格科技股份有限公司 | Waterlogging prediction method, waterlogging prediction system and storage medium |
| CN115798151A (en) * | 2022-11-16 | 2023-03-14 | 中国科学院新疆生态与地理研究所 | Urban flood risk monitoring system and method |
-
2023
- 2023-09-04 CN CN202311129638.0A patent/CN116861317A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222427A (en) * | 2019-06-10 | 2019-09-10 | 河海大学 | A kind of analysis method based on mathematical model urban waterlogging |
| CN111582541A (en) * | 2020-03-27 | 2020-08-25 | 合肥学院 | Firefly algorithm-based inland inundation model prediction method |
| CN113344305A (en) * | 2021-08-05 | 2021-09-03 | 中国水利水电科学研究院 | Rapid prediction method for rainstorm waterlogging event |
| CN114048943A (en) * | 2021-10-08 | 2022-02-15 | 广州珠矶科技有限公司 | Urban waterlogging analysis method, device and medium based on unmanned aerial vehicle BIM and SWMM |
| CN114254561A (en) * | 2021-12-17 | 2022-03-29 | 奥格科技股份有限公司 | Waterlogging prediction method, waterlogging prediction system and storage medium |
| CN115798151A (en) * | 2022-11-16 | 2023-03-14 | 中国科学院新疆生态与地理研究所 | Urban flood risk monitoring system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107831285B (en) | A kind of dystrophication monitoring system and method based on Internet of Things | |
| Wang et al. | Developing an early-warning system for air quality prediction and assessment of cities in China | |
| CN111652777B (en) | Flood emergency danger avoiding method | |
| WO2016034142A1 (en) | Smart allocation method for electric emergency response supplies | |
| CN108320462A (en) | Urban Flood control early-warning and predicting system | |
| CN106128035A (en) | The geological disaster forecasting method merged based on neutral net and multi-parameter information | |
| CN105004396A (en) | Water level monitoring method | |
| CN104992245A (en) | Generalized-entropy-theory-based dynamic intelligent comprehensive analysis method for water environment risk management | |
| CN112711840B (en) | Watershed sudden water pollution tracing method based on cloud edge cooperation | |
| CN104408900A (en) | Dynamic optimization based neural network flood warning device and method | |
| CN107516167A (en) | Municipal drainage integration operation intelligent distribution and management system | |
| CN115063020B (en) | Multi-dimensional safety scheduling device and method for cascade hydropower station based on risk monitoring fusion | |
| CN112668846A (en) | Intelligent ship lock scheduling system | |
| CN109587219A (en) | The setting and management method at Locating Type city Internet of Things edge calculations point station | |
| CN106801463A (en) | The hull-borne detection method and system of drainpipe | |
| CN106845780B (en) | System and method for analyzing operating conditions of generator set | |
| Ye et al. | The system architecture of smart water grid for water security | |
| CN207924874U (en) | Urban Flood control early-warning and predicting device | |
| CN114877933A (en) | Urban road accumulated water monitoring and early warning method and system | |
| CN104281983A (en) | Method and system for dispatching resources needed by emergency repair of power distribution network | |
| CN104346695A (en) | Emergency dispatching circumstance generation method for water pollution emergency | |
| CN109272254A (en) | A kind of in line comprehensive monitor system of sewage zero | |
| CN114049747A (en) | GSM technology-based urban drainage remote monitoring system | |
| CN116664095B (en) | Security inspection prediction method based on deep learning | |
| CN109816978A (en) | Consider the region group system for traffic guiding and method of driver dynamic respondent behavior |
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 |