CN115062675A - Full-spectrum pollution tracing method based on neural network and cloud system - Google Patents
Full-spectrum pollution tracing method based on neural network and cloud system Download PDFInfo
- Publication number
- CN115062675A CN115062675A CN202210901270.4A CN202210901270A CN115062675A CN 115062675 A CN115062675 A CN 115062675A CN 202210901270 A CN202210901270 A CN 202210901270A CN 115062675 A CN115062675 A CN 115062675A
- Authority
- CN
- China
- Prior art keywords
- data
- full
- module
- model
- spectrum
- 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
- 238000001228 spectrum Methods 0.000 title claims abstract description 125
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 24
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 92
- 238000012544 monitoring process Methods 0.000 claims abstract description 68
- 238000004458 analytical method Methods 0.000 claims abstract description 21
- 238000013524 data verification Methods 0.000 claims abstract description 16
- 238000012937 correction Methods 0.000 claims abstract description 12
- 238000011835 investigation Methods 0.000 claims abstract description 5
- 238000012795 verification Methods 0.000 claims abstract description 4
- 238000003062 neural network model Methods 0.000 claims description 21
- 238000012360 testing method Methods 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 17
- 238000010276 construction Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 8
- 230000000007 visual effect Effects 0.000 claims description 8
- 230000003139 buffering effect Effects 0.000 claims description 6
- 238000007639 printing Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000001506 fluorescence spectroscopy Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000003911 water pollution Methods 0.000 description 1
Images
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Biophysics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention discloses a full-spectrum pollution tracing method and a cloud system based on a neural network, which relate to the technical field of water quality detection and comprise a model building module, a data tracing module, a data verification module and a signal monitoring module; the model building module is used for building a full spectrum pollution traceability model based on a neural network according to the full spectrum water quality characteristic database; the data tracing module is used for acquiring full-spectrum water quality monitoring data collected by the data sensing module, inputting the full-spectrum water quality monitoring data into the full-spectrum pollution tracing model for automatic tracing analysis, and outputting suspected pollution sources and similarity; the pollution source can be traced quickly, economically, intelligently and accurately; the data verification module is used for carrying out investigation and verification on suspected pollution sources and similarity output by the data tracing module and judging whether the tracing is qualified or not; the signal monitoring module is used for monitoring unqualified signals and performing correction analysis on the full-spectrum pollution traceability model so as to improve the traceability accuracy of the corresponding model.
Description
Technical Field
The invention relates to the technical field of water quality detection, in particular to a full-spectrum pollution tracing method based on a neural network and a cloud system.
Background
The problem of the watershed water environment pollution gradually becomes a prominent problem restricting the sustainable development of society and economy, and becomes the focus of public attention. The conventional pollution accident source tracing method mainly comprises the steps of carrying out post-manual on-site investigation and tracing, and is large in workload and insufficient in timeliness, so that the discharge path and the pollution source can be determined for several days, and the most precious pollution control time at the initial stage of the pollution accident is lost.
In the aspect of tracing water pollution, the main methods at home and abroad comprise a PMF analytic method, a three-dimensional fluorescence spectrometry, an isotope method and the like. The PMF analysis method is based on a PMF analysis model to realize the identification of pollution types, and cannot realize the pollution source locking. The three-dimensional fluorescence tracing technology has visual images and rich contained information, is one of effective tracing means, usually directly locks a pollution source, needs to supplement water quality three-dimensional fluorescence monitoring under the condition of meeting the conventional water quality monitoring, and causes high tracing cost. The isotope method tracing technology usually adopts isotopes to trace after a pollution event occurs, and the problem of insufficient timeliness exists. At present, the application of tracing analysis by using full spectrum data is rare in the prior art, and the tracing is untimely, difficult and high in cost aiming at sudden water environmental pollution, a pollution path is not clear, and a pollution source cannot be caught; the pollution source is difficult to be locked economically, accurately and quickly; based on the defects, the invention provides a full-spectrum pollution tracing method based on a neural network and a cloud system.
Disclosure of Invention
The present invention is directed to solving at least one of the problems of the prior art. Therefore, the invention provides a full spectrum pollution tracing method and a cloud system based on a neural network; full spectrum data based on water quality conventional monitoring does not need other supplementary monitoring, a traceability model is established through a neural network, and rapid, economic, intelligent and accurate traceability of a pollution source is realized through comparison and analysis with full spectrum water quality characteristic data.
In order to achieve the above object, an embodiment according to a first aspect of the present invention provides a full spectrum pollution traceability cloud system based on a neural network, including a model building module, a data sensing module, a data verification module, and a signal monitoring module;
the model construction module is used for constructing a full spectrum pollution traceability model based on a neural network according to a full spectrum water quality characteristic database; feeding back the successfully constructed full spectrum pollution traceability model to the data traceability module; the data tracing module is used for acquiring full-spectrum water quality monitoring data collected by the data sensing module; inputting full spectrum water quality monitoring data into a full spectrum pollution traceability model for automatic traceability analysis, and outputting suspected pollution sources and similarity;
the data verification module is used for carrying out investigation and verification on the suspected pollution source and the similarity output by the data tracing module, comparing the suspected pollution source with the real pollution source and judging whether the tracing is qualified or not;
the signal monitoring module is connected with the data verification module and is used for monitoring unqualified signals, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and calculating to obtain the traceability deviation value PL of the corresponding model; and judging whether the corresponding model needs to correct the relevant parameters.
Further, the specific analysis steps of the signal monitoring module are as follows:
counting the occurrence frequency of unqualified signals to be P1 in a preset time period; intercepting a time period between adjacent unqualified signals as a deviation buffer period; counting the occurrence frequency of qualified signals in each deviation buffering time period as deviation buffering frequency Li; comparing Li to a buffer frequency threshold;
counting the number of times that Li is smaller than a buffer frequency threshold value as P2, when Li is smaller than the buffer frequency threshold value, obtaining a difference value between Li and the buffer frequency threshold value and summing the difference value to obtain a difference and buffer value CH, and calculating by using a formula CS = P2 × g1+ CH × g2 to obtain a difference and buffer coefficient CS, wherein g1 and g2 are coefficient factors;
using formulasCalculating to obtain a traceability deviation value PL of the corresponding model, wherein g3 and g4 are coefficient factors; comparing the tracing deviation value PL with a preset deviation value threshold; if PL is larger than or equal to the preset bias value threshold value, the tracing result error of the corresponding model is judged to be larger, and a correction signal is generated.
Furthermore, the signal monitoring module is used for transmitting the correction signal to the model correction module through the cloud platform so as to remind a manager to correct relevant parameters of the full-spectrum pollution traceability model, and performing iterative optimization on the full-spectrum pollution traceability model by combining traceability errors.
Further, the specific construction steps of the model construction module are as follows:
s1: collecting water samples of different pollution sources, different industries and different water bodies, scanning the water samples by adopting a full spectrum, and establishing a full spectrum water quality characteristic database;
s2: visualizing the original coordinate data of the full spectrum water quality characteristic database and converting the visualized original coordinate data into a water quality full spectrogram with consistent coordinate axes and the same size;
s3: decoding the water quality full spectrogram and converting the water quality full spectrogram into characteristic matrix data for computer identification; standardizing the decoded characteristic matrix data, extracting a pollution source name or an industry type or a water body type of each picture as an identification label, and integrating the pollution source name, the industry type or the water body type into a data set containing a water quality full spectrum characteristic matrix and the label;
s4: building a 4-layer neural network model based on Tensorflow, wherein the input layer is 1 layer, the hidden layer is 2 layers, and the output layer is 1 layer;
s5: randomly mixing the data sets according to rows, and dividing the data sets into a training set and a testing set according to a set proportion; training and testing the neural network model through the training set and the testing set, and marking the tested neural network model as a full spectrum pollution traceability model.
Further, the data perception module is used for collecting full spectrum water quality monitoring data, wherein the data source form comprises online monitoring, remote access and cloud user uploading.
Further, the system also comprises a data display module; the data display module is connected with the data traceability module and is used for realizing the visual display of the full-spectrum water quality monitoring data and the traceability result.
Further, the system also comprises a data query module; and the data query module is used for a user to input keywords through the mobile phone terminal to query the traceability result of the corresponding polluted water sample.
Further, a full spectrum pollution tracing method based on a neural network comprises the following steps:
the method comprises the following steps: constructing a full spectrum pollution traceability model based on a neural network according to a full spectrum water quality characteristic database through a model construction module;
step two: a user logs in a water environment tracing SaaS cloud system, and uploads full-spectrum water quality monitoring data of a polluted water sample to a data tracing module through a data sensing module;
step three: the data traceability module automatically conducts traceability analysis by inputting full-spectrum water quality monitoring data into the full-spectrum pollution traceability model, outputs suspected pollution sources and similarity, and realizes visual display of the full-spectrum water quality monitoring data and traceability results through the data display module;
step four: the output suspected pollution source and the similarity are investigated and verified through a data verification module to obtain a source tracing error; judging whether the tracing is qualified or not;
step five: and monitoring the unqualified signals through a signal monitoring module, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and judging whether the corresponding model needs to be corrected and carrying out iterative optimization.
Compared with the prior art, the invention has the beneficial effects that:
1. the model construction module is used for constructing a full spectrum pollution traceability model based on a neural network according to a full spectrum water quality characteristic database; the data tracing module is used for acquiring full-spectrum water quality monitoring data collected by the data sensing module, inputting the full-spectrum water quality monitoring data into the full-spectrum pollution tracing model for automatic tracing analysis, and outputting suspected pollution sources and similarity; the method is based on full spectrum data of conventional water quality monitoring, does not need other supplementary monitoring, establishes a traceability model through a neural network, and realizes quick, economic, intelligent and accurate traceability of a pollution source through comparison and analysis with full spectrum water quality characteristic data;
2. the data verification module is used for investigating and verifying suspected pollution sources and similarity output by the data tracing module to obtain tracing errors; if the tracing error is within the allowable range, generating a qualified signal; otherwise, generating an unqualified signal; the signal monitoring module is used for monitoring unqualified signals, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and calculating to obtain the traceability deviation value PL of the corresponding model; if PL is larger than or equal to a preset bias value threshold value, judging that the error of the tracing result of the corresponding model is larger, and generating a correction signal; the management personnel are reminded to correct the relevant parameters of the full-spectrum pollution traceability model, iterative optimization is carried out on the full-spectrum pollution traceability model by combining traceability errors, and the traceability precision of the corresponding model is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a system block diagram of a full spectrum pollution tracing cloud system based on a neural network according to the present invention.
Fig. 2 is a flow chart of a full spectrum pollution tracing method based on a neural network according to the present invention.
FIG. 3 is a flow chart of the construction of the full spectrum pollution traceability model in the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments, and it should be understood 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.
As shown in fig. 1 to 3, a full spectrum pollution traceability cloud system based on a neural network includes a model building module, a data sensing module, a data traceability module, a data display module, a data query module, a cloud platform, a data verification module, a signal monitoring module, and a model modification module;
the model building module is used for building a full spectrum pollution traceability model based on a neural network and feeding back the successfully built full spectrum pollution traceability model to the data traceability module; the concrete construction steps are as follows:
s1: collecting water samples of different pollution sources, different industries and different water bodies, scanning the water samples by adopting a full spectrum (200-850 nm), and establishing a full spectrum water quality characteristic database of the different pollution sources, the different industries and the different water bodies;
s2: visualizing the original coordinate data of the full spectrum water quality characteristic database, and converting the visualized original coordinate data into a water quality full spectrogram with consistent coordinate axes and the same size;
s3: decoding the water quality full spectrogram, and converting the water quality full spectrogram into feature matrix data which can be recognized by a computer; the water quality full spectrum of each enterprise contains 100-150 ten thousand characteristics, the decoded characteristic matrix data is standardized, the pollution source name or industry type or water body type of each picture is extracted as an identification label, and the identification label is integrated into a data set containing a water quality full spectrum characteristic matrix and the label;
s4: building a 4-layer neural network model based on Tensorflow, wherein the input layer is 1 layer, the hidden layer is 2 layers, and the output layer is 1 layer;
s5: randomly mixing the data sets according to rows, and dividing the data sets into a training set and a testing set according to a set proportion; training and testing the neural network model through a training set and a testing set, and marking the tested neural network model as a full spectrum pollution traceability model; the method comprises the following specific steps:
training the neural network model by using the divided training set data, printing the training accuracy, continuously performing model parameter adjustment, and optimizing the recognition accuracy of the neural network model; when the model identification accuracy of the training set data is more than 90%, testing the neural network model by using the test set data;
adjusting model parameters during testing of the neural network model; when the model identification accuracy of the test set data is more than 85%, the neural network model is basically tested; marking the tested neural network model as a full spectrum pollution traceability model;
the data perception module is used for collecting full-spectrum water quality monitoring data, wherein the data source form comprises online monitoring, remote access, cloud user uploading and the like;
the data tracing module is connected with the data sensing module and is used for acquiring full-spectrum water quality monitoring data collected by the data sensing module; inputting full-spectrum water quality monitoring data into a full-spectrum pollution traceability model to automatically perform traceability analysis, and outputting suspected pollution sources and similarity;
the data display module is connected with the data traceability module and is used for realizing the visual display of the full-spectrum water quality monitoring data and the traceability result; the data query module is connected with the data display module and is used for a user to input keywords through the mobile phone terminal to query the tracing result of the corresponding polluted water sample;
the method is based on full spectrum data of conventional water quality monitoring, does not need other supplementary monitoring, establishes a traceability model through a neural network, and realizes quick, economic, intelligent and accurate traceability of a pollution source through comparison and analysis with full spectrum water quality characteristic data;
the data verification module is connected with the data tracing module and is used for investigating and verifying the suspected pollution source and the similarity output by the data tracing module and comparing the suspected pollution source with the real pollution source to obtain a tracing error; if the tracing error is within the allowable range, generating a qualified signal; otherwise, generating an unqualified signal;
the signal monitoring module is connected with the data verification module and is used for monitoring unqualified signals, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and judging whether the corresponding model needs to correct related parameters or not so as to improve the traceability accuracy of the model; the specific analysis steps are as follows:
counting the occurrence frequency of unqualified signals to be P1 in a preset time period; intercepting a time period between adjacent unqualified signals as a deviation buffer period; counting the occurrence frequency of qualified signals in each deviation buffering time period as deviation buffering frequency Li; comparing Li to a buffer frequency threshold;
counting the number of times that Li is smaller than a buffer frequency threshold value as P2, when Li is smaller than the buffer frequency threshold value, obtaining a difference value between Li and the buffer frequency threshold value and summing the difference value to obtain a difference and buffer value CH, and calculating by using a formula CS = P2 × g1+ CH × g2 to obtain a difference and buffer coefficient CS, wherein g1 and g2 are coefficient factors;
using formulasCalculating to obtain a tracing deviation value PL of the corresponding model, wherein g3 and g4 are coefficient factors; comparing the tracing deviation value PL with a preset deviation value threshold; if PL is larger than or equal to a preset bias value threshold value, judging that the error of the tracing result of the corresponding model is larger, and generating a correction signal;
the signal monitoring module is used for transmitting the correction signal to the model correction module through the cloud platform so as to remind a manager of correcting relevant parameters of the full-spectrum pollution traceability model, iterative optimization is carried out on the full-spectrum pollution traceability model by combining traceability errors, and the traceability accuracy of the corresponding model is improved.
A full-spectrum pollution tracing method based on a neural network is applied to the full-spectrum pollution tracing cloud system based on the neural network, and comprises the following steps:
the method comprises the following steps: constructing a full spectrum pollution traceability model based on a neural network according to a full spectrum water quality characteristic database through a model construction module;
step two: a user logs in a water environment tracing SaaS cloud system, and uploads full-spectrum water quality monitoring data of a polluted water sample to a data tracing module through a data sensing module;
step three: the data traceability module automatically conducts traceability analysis by inputting full-spectrum water quality monitoring data into the full-spectrum pollution traceability model, outputs suspected pollution sources and similarity, and realizes visual display of the full-spectrum water quality monitoring data and traceability results through the data display module;
step four: the output suspected pollution source and the similarity are investigated and verified through a data verification module to obtain a source tracing error; judging whether the tracing is qualified or not;
step five: and monitoring the unqualified signals through a signal monitoring module, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and judging whether the corresponding model needs to be corrected and carrying out iterative optimization.
The above formulas are all calculated by removing dimensions and taking numerical values thereof, the formula is a formula which is obtained by acquiring a large amount of data and performing software simulation to obtain the closest real situation, and the preset parameters and the preset threshold value in the formula are set by the technical personnel in the field according to the actual situation or obtained by simulating a large amount of data.
The working principle of the invention is as follows:
when the full-spectrum pollution traceability method and the cloud system based on the neural network work, a model construction module is used for constructing a full-spectrum pollution traceability model based on the neural network according to a full-spectrum water quality characteristic database and feeding the full-spectrum pollution traceability model back to a data traceability module; the data tracing module is used for acquiring full-spectrum water quality monitoring data collected by the data sensing module; inputting full spectrum water quality monitoring data into a full spectrum pollution traceability model for automatic traceability analysis, and outputting suspected pollution sources and similarity; the data display module is used for realizing the visual display of the full-spectrum water quality monitoring data and the tracing result; the method is based on full spectrum data of conventional water quality monitoring, does not need other supplementary monitoring, establishes a traceability model through a neural network, and realizes quick, economic, intelligent and accurate traceability of a pollution source through comparison and analysis with full spectrum water quality characteristic data;
the data verification module is used for carrying out investigation and verification on the suspected pollution source and the similarity output by the data tracing module and comparing the suspected pollution source with the real pollution source to obtain a tracing error; if the tracing error is within the allowable range, generating a qualified signal; otherwise, generating an unqualified signal; the signal monitoring module is used for monitoring unqualified signals, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and calculating to obtain the traceability deviation value PL of the corresponding model; if PL is larger than or equal to a preset bias value threshold value, judging that the error of the tracing result of the corresponding model is larger, and generating a correction signal; the method can remind management personnel to correct relevant parameters of the full-spectrum pollution traceability model, iterative optimization is carried out on the full-spectrum pollution traceability model by combining traceability errors, and the traceability accuracy of the corresponding model is improved.
In the description herein, references to the description of "one embodiment," "an example," "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (6)
1. The full-spectrum pollution source tracing cloud system based on the neural network is characterized by comprising a model building module, a data sensing module, a data verification module and a signal monitoring module;
the model building module is used for building a full spectrum pollution traceability model based on a neural network and feeding back the successfully built full spectrum pollution traceability model to the data traceability module; the concrete construction steps are as follows:
s1: collecting water samples of different pollution sources, different industries and different water bodies, scanning the water samples by adopting a full spectrum, and establishing a full spectrum water quality characteristic database;
s2: visualizing the original coordinate data of the full spectrum water quality characteristic database, and converting the visualized original coordinate data into a water quality full spectrogram with consistent coordinate axes and the same size;
s3: decoding the water quality full spectrogram, and converting the water quality full spectrogram into feature matrix data which can be recognized by a computer; standardizing the decoded characteristic matrix data, extracting the pollution source name or industry type or water body type of each picture as an identification label, and integrating the pollution source name or industry type or water body type into a data set containing a water quality full spectrum characteristic matrix and the label;
s4: building a 4-layer neural network model based on Tensorflow, wherein the input layer is 1 layer, the hidden layer is 2 layers, and the output layer is 1 layer;
s5: randomly mixing the data sets according to rows, and dividing the data sets into a training set and a testing set according to a set proportion; training and testing the neural network model through a training set and a testing set, and marking the tested neural network model as a full spectrum pollution traceability model; the method specifically comprises the following steps:
training the neural network model by using the divided training set data, printing the training accuracy, continuously performing model parameter adjustment, and optimizing the recognition accuracy of the neural network model; when the model identification accuracy of the training set data is more than 90%, testing the neural network model by using the test set data;
adjusting model parameters during testing of the neural network model; when the model identification accuracy of the test set data is more than 85%, the neural network model is basically tested; marking the tested neural network model as a full spectrum pollution traceability model;
the data tracing module is used for acquiring full-spectrum water quality monitoring data collected by the data sensing module; inputting full-spectrum water quality monitoring data into a full-spectrum pollution traceability model to automatically perform traceability analysis, and outputting suspected pollution sources and similarity;
the data verification module is used for carrying out investigation and verification on the suspected pollution source and the similarity output by the data tracing module, comparing the suspected pollution source with the real pollution source and judging whether the tracing is qualified or not;
the signal monitoring module is connected with the data verification module and is used for monitoring unqualified signals, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and calculating to obtain the traceability deviation value PL of the corresponding model; judging whether the corresponding model needs to correct the relevant parameters;
the signal monitoring module is used for transmitting the correction signal to the model correction module through the cloud platform so as to remind a manager to correct relevant parameters of the full-spectrum pollution traceability model, and iteration optimization is carried out on the full-spectrum pollution traceability model by combining traceability errors.
2. The full-spectrum pollution traceability cloud system based on the neural network as claimed in claim 1, wherein the specific analysis steps of the signal monitoring module are as follows:
counting the occurrence frequency of unqualified signals to be P1 in a preset time period; intercepting a time period between adjacent unqualified signals as a deviation buffer period; counting the occurrence frequency of qualified signals in each deviation buffering time period as deviation buffering frequency Li; comparing Li to a buffer frequency threshold;
counting the number of times that Li is smaller than a buffer frequency threshold value as P2, when Li is smaller than the buffer frequency threshold value, obtaining a difference value between Li and the buffer frequency threshold value and summing the difference value to obtain a difference and buffer value CH, and calculating by using a formula CS = P2 × g1+ CH × g2 to obtain a difference and buffer coefficient CS, wherein g1 and g2 are coefficient factors;
using a formulaCalculating to obtain a traceability deviation value PL of the corresponding model, wherein g3 and g4 are coefficient factors; comparing the tracing deviation value PL with a preset deviation value threshold; if PL is larger than or equal to the preset bias value threshold value, the tracing result error of the corresponding model is judged to be larger, and a correction signal is generated.
3. The full spectrum pollution traceability cloud system based on a neural network as claimed in claim 1, wherein the data sensing module is configured to collect full spectrum water quality monitoring data, wherein the data source form comprises online monitoring, remote access and cloud user uploading.
4. The full spectrum pollution traceability cloud system based on a neural network as claimed in claim 1, further comprising a data presentation module; the data display module is connected with the data traceability module and is used for realizing the visual display of the full-spectrum water quality monitoring data and the traceability result.
5. The full spectrum pollution traceability cloud system based on the neural network as claimed in claim 4, further comprising a data query module; and the data query module is used for a user to input keywords through the mobile phone terminal to query the traceability result of the corresponding polluted water sample.
6. The full-spectrum pollution traceability method based on the neural network is applied to the full-spectrum pollution traceability cloud system based on the neural network as claimed in any one of claims 1 to 5, and is characterized by comprising the following steps:
the method comprises the following steps: constructing a full spectrum pollution traceability model based on a neural network according to a full spectrum water quality characteristic database through a model construction module;
step two: a user logs in a water environment tracing SaaS cloud system, and uploads full-spectrum water quality monitoring data of a polluted water sample to a data tracing module through a data sensing module;
step three: the data traceability module automatically conducts traceability analysis by inputting full-spectrum water quality monitoring data into the full-spectrum pollution traceability model, outputs suspected pollution sources and similarity, and realizes visual display of the full-spectrum water quality monitoring data and traceability results through the data display module;
step four: the output suspected pollution source and the similarity are investigated and verified through a data verification module to obtain a source tracing error; judging whether the tracing is qualified or not;
step five: and monitoring the unqualified signals through a signal monitoring module, correcting and analyzing the full-spectrum pollution traceability model according to the monitored unqualified signals, and judging whether the corresponding model needs to be corrected and carrying out iterative optimization.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210901270.4A CN115062675A (en) | 2022-07-28 | 2022-07-28 | Full-spectrum pollution tracing method based on neural network and cloud system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210901270.4A CN115062675A (en) | 2022-07-28 | 2022-07-28 | Full-spectrum pollution tracing method based on neural network and cloud system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115062675A true CN115062675A (en) | 2022-09-16 |
Family
ID=83205982
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210901270.4A Pending CN115062675A (en) | 2022-07-28 | 2022-07-28 | Full-spectrum pollution tracing method based on neural network and cloud system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115062675A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115830068A (en) * | 2022-11-29 | 2023-03-21 | 中国环境科学研究院 | Pollution tracing big data model based on pollution path identification |
CN116008495A (en) * | 2022-12-28 | 2023-04-25 | 廊坊卓筑建筑工程有限公司 | Water body data acquisition and analysis system and method for surface water |
CN116011317A (en) * | 2022-11-29 | 2023-04-25 | 北京工业大学 | Small-scale near-real-time atmospheric pollution tracing method based on multi-method fusion |
CN116297251A (en) * | 2023-05-17 | 2023-06-23 | 安徽新宇环保科技股份有限公司 | Multi-sensor combined water quality detection system and detection probe thereof |
-
2022
- 2022-07-28 CN CN202210901270.4A patent/CN115062675A/en active Pending
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115830068A (en) * | 2022-11-29 | 2023-03-21 | 中国环境科学研究院 | Pollution tracing big data model based on pollution path identification |
CN116011317A (en) * | 2022-11-29 | 2023-04-25 | 北京工业大学 | Small-scale near-real-time atmospheric pollution tracing method based on multi-method fusion |
CN115830068B (en) * | 2022-11-29 | 2023-06-20 | 中国环境科学研究院 | Pollution tracing big data model based on pollution path identification |
CN116011317B (en) * | 2022-11-29 | 2023-12-08 | 北京工业大学 | Small-scale near-real-time atmospheric pollution tracing method based on multi-method fusion |
CN116008495A (en) * | 2022-12-28 | 2023-04-25 | 廊坊卓筑建筑工程有限公司 | Water body data acquisition and analysis system and method for surface water |
CN116297251A (en) * | 2023-05-17 | 2023-06-23 | 安徽新宇环保科技股份有限公司 | Multi-sensor combined water quality detection system and detection probe thereof |
CN116297251B (en) * | 2023-05-17 | 2023-08-29 | 安徽新宇环保科技股份有限公司 | Multi-sensor combined water quality detection system and detection probe thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115062675A (en) | Full-spectrum pollution tracing method based on neural network and cloud system | |
CN107862338B (en) | Marine environment monitoring data quality management method and system based on double inspection method | |
CN113407517B (en) | Data quality health degree analysis method and system based on multidimensional analysis technology | |
CN111766253A (en) | Solder paste printing quality detection method, data processing device, and computer storage medium | |
CN103528617A (en) | Cockpit instrument automatic recognizing and detecting method and device | |
CN114168906A (en) | Mapping geographic information data acquisition system based on cloud computing | |
US20230014095A1 (en) | Method and system for recognizing environmental protection equipment based on deep hierarchical fuzzy algorithm | |
CN115097796B (en) | Quality control system and method for simulating big data and correcting AQL value | |
CN114625901B (en) | Multi-algorithm integration method and device | |
CN115409395A (en) | Quality acceptance inspection method and system for hydraulic construction engineering | |
CN113570330A (en) | System and method for evaluating simulation training effect of emergency environment | |
CN112966921A (en) | Innovation ability evaluation method based on programming ability evaluation | |
CN115481941B (en) | Multifunctional area combined intelligent security management method and system | |
CN117217020A (en) | Industrial model construction method and system based on digital twin | |
CN116681556A (en) | Intelligent remote teaching cloud platform system and method based on education big data | |
CN116307858A (en) | Mapping result quality inspection and scoring system | |
CN113706178B (en) | Supervision method for product inspection process | |
CN112505337B (en) | Data processing method for assisting analysis of sample | |
CN112329108A (en) | Optimized anti-floating checking calculation method and system for subway station | |
CN117076454B (en) | Engineering quality acceptance form data structured storage method and system | |
CN113609643B (en) | Time acquisition method and system for digital operation basic action unit of nuclear power station | |
CN117808366A (en) | Digital model system for black land protection utilization and safety productivity evaluation | |
CN116842372A (en) | Deep learning sample labeling method and system based on online education big data | |
CN115391080A (en) | Process consistency checking method and system based on attribute filtering | |
CN116932679A (en) | GIS technology-based multi-mode industrial weather service inspection application method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |