CN107664680A - A kind of adaptive acquiring method and device of water quality soft-sensing model - Google Patents
A kind of adaptive acquiring method and device of water quality soft-sensing model Download PDFInfo
- Publication number
- CN107664680A CN107664680A CN201610601948.1A CN201610601948A CN107664680A CN 107664680 A CN107664680 A CN 107664680A CN 201610601948 A CN201610601948 A CN 201610601948A CN 107664680 A CN107664680 A CN 107664680A
- Authority
- CN
- China
- Prior art keywords
- soft
- sensing model
- water quality
- model
- sensing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/18—Water
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
Embodiment of the present invention is related to fields of measurement, discloses a kind of adaptive acquiring method and device of water quality soft-sensing model.In the present invention, water quality soft-sensing model adaptive acquiring method, including:Multiple water quality soft-sensing models that prestore form water quality soft-sensing model storehouse, and characteristic information is added for each water quality soft-sensing model;Receive model feature information to be selected;Using the characteristic information of reception, from water quality soft-sensing model storehouse, according to the similarity of the characteristic information of reception and each water quality soft-sensing model characteristic information, soft-sensing model is matched;Soft-sensing model is tested, the degree of accuracy of test not super preset value when, soft-sensing model parameter is adjusted, until the degree of accuracy of test is to preset value;The degree of accuracy tested is reached into the soft-sensing model of preset value as the water quality soft-sensing model adaptively obtained.The present invention discloses a kind of adaptive acquisition device of water quality soft-sensing model.Embodiment of the present invention accelerates the formation speed of soft-sensing model, while reduces cost.
Description
Technical field
The present invention relates to field of measuring technique, more particularly to a kind of adaptive acquiring technology of water quality soft-sensing model.
Background technology
In maintaining the existence of people, ensureing economic construction and safeguard all physical features of social development, the importance of water
Needless to say.At present, with industrial expansion, the increase of population, the aggravation of urbanization and chemical fertilizer, the increase of Pesticide use amount,
Water as Source of life has received serious pollution.Water pollution reduces the use function of water body, exacerbates water resource
Shortage;Water pollution heavy damage ecological environment, influence human survival.Moreover, the water that people drink usually, it is also possible to receive multi-party
The pollution in face, therefore, the detection of water quality, gradually paid attention to by everybody.
With attention of the country to industry, water quality monitoring industry size constantly expands, wherein water quality monitoring equipment production enterprise
Industry quantity increases rapidly, when needing to measure water quality in the prior art, first has to generate a new water quality soft-sensing model, this is just
Need to expend huge fund.In new water quality soft-sensing model is generated, it is also necessary to substantial amounts of historical data, pass through history number
According to training and generate new measurement model.Wherein, historical data not only needs directly to be obtained by equipment such as sensors, it is also necessary to tired
A very long time is counted, this will result in the consuming of time.
The content of the invention
The purpose of embodiment of the present invention is to provide a kind of adaptive acquiring method and device of water quality soft-sensing model,
The formation speed of soft-sensing model is accelerated, while reduces cost.
Obtained in order to solve the above technical problems, embodiments of the present invention provide a kind of the adaptive of water quality soft-sensing model
Method is taken, including:
Multiple water quality soft-sensing models that prestore form water quality soft-sensing model storehouse, and feature is added for each water quality soft-sensing model
Information;
Receive the characteristic information of model to be selected;
Received characteristic information is utilized, from water quality soft-sensing model storehouse, according to received characteristic information and respectively
The similarity of the characteristic information of water quality soft-sensing model, match the first soft-sensing model;
First soft-sensing model is tested, when the degree of accuracy tested is no more than preset value, adjusts the first soft survey
The parameter of model is measured, until the degree of accuracy tested reaches preset value;
The degree of accuracy tested is reached into the first soft-sensing model of preset value as the soft survey of the water quality adaptively got
Measure model.
Embodiments of the present invention additionally provide a kind of adaptive acquisition device of water quality soft-sensing model, including:
Memory module, for prestoring, multiple water quality soft-sensing models form water quality soft-sensing model storehouse, are the soft survey of each water quality
Measure model addition characteristic information;
Communication module, for receiving the characteristic information of model to be selected;
Processing module, for utilizing received characteristic information, from water quality soft-sensing model storehouse, according to received
The similarity of the characteristic information of characteristic information and each water quality soft-sensing model, matches the first soft-sensing model;
Test module, for testing the first soft-sensing model, when the degree of accuracy tested is no more than preset value,
The parameter of the first soft-sensing model is adjusted, until the degree of accuracy tested reaches preset value;
Output module, the first soft-sensing model for the degree of accuracy tested to be reached to preset value obtain as adaptive
The water quality soft-sensing model arrived.
Embodiment of the present invention in terms of existing technologies, utilizes existing water quality soft-sensing model and the water quality received
The similarity mode of soft-sensing model, the first closest soft-sensing model is found out, and using constantly test, be localized
The accommodation of parameter, so as to which quick obtaining meets the soft-sensing model of water quality to be measured, greatly accelerate water quality hard measurement
The formation speed of model, new water quality soft-sensing model is regenerated so as to avoid, reduces the manufacturing cost of model.
In addition, also include:By the water quality soft-sensing model adaptively got deposit water quality soft-sensing model storehouse.Constantly more
New water quality soft-sensing model storehouse, can expand water quality soft-sensing model quantity, accurate with Model Matching to be selected so as to be easier to find
The high water quality soft-sensing model of exactness, improve the possibility for finding the model to be selected more met.
In addition, characteristic information is set according to the application environment of water quality soft-sensing model;Received characteristic information and each
The similarity of the characteristic information of water quality soft-sensing model, according to the similar of the application environment of model to be selected and water quality soft-sensing model
Degree determines.Because application environment is similar, it can make it that model to be measured is more close, so comparing the similarity of application environment, so that it may
To find more like model to be measured.
In addition, application environment includes one below or its any combination:Used by water quality index to be measured, water to be measured at water
Geographical position residing for science and engineering skill, water to be measured.Application environment is used as using different parameters so that selection is more diversified, realizes different
The determination of angle similitude, prevent from, because application environment is different, causing comparative result larger deviation occur.
In addition, if application environment includes:Water treatment technology used by water to be measured, then according to technique correlation, it is determined that
The similarity of water treatment technology used by water to be measured.It is defined for how to obtain Process similarity so that technique phase
Become apparent from like degree judgment basis.
In addition, in testing the first soft-sensing model, specifically include:Adopted in the water environment that model to be selected is surveyed
Sample;The first soft-sensing model is tested using institute's water sampling.Sample is tested, can avoid carrying out online reality
When measure, and use off-line measurement mode instead, avoid the equipment using being used needed for on-line measurement as far as possible, further reduce test into
This.
In addition, if the quantity of the first soft-sensing model matched is more than 1, then the first soft-sensing model is carried out
Test, according in the step of parameter of test result the first soft-sensing model of adjustment, specifically include:Selected according to test result accurate
The soft-sensing model of exactness highest first, and adjust the parameter of first soft-sensing model.Allow to match first is soft
Measurement model quantity more than one, so as to therefrom select a soft-sensing model of degree of accuracy highest first, and it is joined
Number is adjusted.
Brief description of the drawings
Fig. 1 is a kind of stream of the adaptive acquiring method of water quality soft-sensing model in first embodiment of the invention
Cheng Tu;
Fig. 2 is a kind of stream of the adaptive acquiring method of water quality soft-sensing model in four embodiment of the invention
Cheng Tu;
Fig. 3 is a kind of knot of the adaptive acquisition device of water quality soft-sensing model in fifth embodiment of the invention
Structure schematic diagram.
Embodiment
To make the purpose, technical scheme and advantage of embodiment of the present invention clearer, below in conjunction with accompanying drawing to this hair
Bright each embodiment is explained in detail.However, it will be understood by those skilled in the art that in each implementation of the invention
In mode, in order that reader more fully understands the application and proposes many ins and outs.It is but even if thin without these technologies
Section and many variations based on following embodiment and modification, can also realize the application technical scheme claimed.
The first embodiment of the present invention is related to a kind of adaptive acquiring method of water quality soft-sensing model.Its flow is as schemed
It is specific as follows shown in 1:
Step 101:Prestore multiple water quality soft-sensing models, and adds characteristic information.
Specifically, multiple water quality soft-sensing models that prestore form water quality soft-sensing model storehouse, are each water quality hard measurement mould
Type adds characteristic information.Prestore multiple water quality soft-sensing models, and multiple water quality soft-sensing models can be frequent from the prior art
Using to water quality soft-sensing model in obtain, and these models are prestored into water quality soft-sensing model storehouse.Meanwhile also need for
Each water quality soft-sensing model addition characteristic information being added in water quality soft-sensing model storehouse, this feature information can be according to water quality
The information such as the design feature of soft-sensing model or water treatment technology to be measured is set, the feature of each water quality soft-sensing model
Information can be different.
It should be noted that the characteristic information of each water quality soft-sensing model can be set according to many kinds of parameters, herein
No longer repeat one by one.
Step 102:Receive the characteristic information of model to be selected.
Specifically, each water quality soft-sensing model being pre-stored in water quality soft-sensing model storehouse, is that existing water quality is soft
Measurement model, when needing to measure water environment, receive the characteristic information of model to be measured.
Step 103:From water quality soft-sensing model storehouse, the first soft-sensing model is matched.
Specifically, received characteristic information is utilized, from water quality soft-sensing model storehouse, according to received feature
The similarity of the characteristic information of information and each water quality soft-sensing model, matches the first soft-sensing model.In a step 102, connect
Model feature information to be measured is received, and in water quality soft-sensing model storehouse, found and high pre- of the characteristic information similarity that is received
Water matter soft-sensing model, and using the water quality soft-sensing model that prestores as the first soft-sensing model.
Step 104:Judge whether the first soft-sensing model degree of accuracy reaches preset value.
Specifically, the first soft-sensing model obtained in step 103 is tested, do not surpassed in the degree of accuracy tested
When crossing preset value, into step 105, otherwise, into step 106.Preset value in present embodiment, can be that user is pre-
The value first set.In this step, the first soft-sensing model degree of accuracy can also be set as a value, if for example,
The one soft-sensing model degree of accuracy is 4, and it is relatively low to represent the degree of accuracy of first soft-sensing model, then, only need to judge that numerical value 4 is
It is no to reach default numerical value.
Step 105:Adjust the first soft-sensing model parameter.
Specifically, if the first soft-sensing model degree of accuracy is not up to preset value, the first soft-sensing model is adjusted
Parameter, the first soft-sensing model after adjusting parameter, is again introduced into step 104, is judged, until the degree of accuracy tested
Reach preset value.
Step 106:By the water quality soft-sensing model adaptively got deposit water quality soft-sensing model storehouse.
Specifically, the degree of accuracy tested is reached into the first soft-sensing model of preset value as adaptively getting
Water quality soft-sensing model.By step 104, the first soft-sensing model degree of accuracy judged reaches preset value, will now obtained
To model as the water quality soft-sensing model adaptively got.Now, got best suit water quality to be measured water quality it is soft
Measurement model, the measurement of water quality to be measured can be actually used in.
In addition, the water quality soft-sensing model adaptively got is stored in water quality soft-sensing model storehouse.So as to expand water
Matter soft-sensing model quantity, when being new one-time detection water quality, it is easier to it is soft to find the water quality high with the Model Matching degree of accuracy to be selected
Measurement model, improve the possibility for finding the model to be selected more met.
In terms of existing technologies, the main distinction and effect are present embodiment:Utilize existing water quality hard measurement mould
The similarity mode of type and the water quality soft-sensing model received, the first closest soft-sensing model is found out, and using not
Disconnected test, the accommodation of parameter is localized, so as to which quick obtaining meets the soft-sensing model of water quality to be measured, greatly
The formation speed of water quality soft-sensing model is accelerated, new water quality soft-sensing model is regenerated so as to avoid, reduces mould
The manufacturing cost of type.
Second embodiment of the present invention is related to a kind of adaptive acquiring method of water quality soft-sensing model.Second embodiment party
Formula is roughly the same with first embodiment, is in place of the main distinction:In second embodiment of the invention, received spy
Reference ceases and the similarity of the characteristic information of each water quality soft-sensing model, according to the application of model to be selected and water quality soft-sensing model
The similarity of environment determines, because application environment is similar, can make it that model to be measured is more close, so comparing the phase of application environment
Like degree, it is possible to find more like model to be measured.
Characteristic information is set according to the application environment of water quality soft-sensing model;Received characteristic information and each water quality are soft
The similarity of the characteristic information of measurement model, it is true according to the similarity of model to be selected and the application environment of water quality soft-sensing model
It is fixed.
Specifically, carried out in the similarity of the characteristic information to receiving and the characteristic information of each water quality soft-sensing model
When comparison, it can be determined according to the similarity of model to be selected and the application environment of water quality soft-sensing model.
Wherein, application environment includes water quality index to be measured, water treatment technology used by water to be measured, geographical residing for water to be measured
One of position or its any combination.For example, can be according to the water quality index similarity of model to be selected and water quality soft-sensing model
It is compared, it is determined that whether the characteristic information received is consistent with the characteristic information of each water quality soft-sensing model.So as to
Different parameters are as application environment so that selection is more diversified, realizes the determination of different angle similitude, prevents due to application
Environment is different, causes comparative result larger deviation occur.
It is noted that if application environment includes:Water treatment technology used by water to be measured is then related according to technique
Property, determine the similarity of water treatment technology used by water to be measured.If application environment includes water treatment technology, according to technique
Correlation, determine its Process similarity.Restriction for how to obtain Process similarity so that Process similarity judgment basis is more
It is clear to add
In terms of existing technologies, the main distinction and effect are present embodiment:Received characteristic information and
The similarity of the characteristic information of each water quality soft-sensing model, according to model to be selected and the phase of the application environment of water quality soft-sensing model
Determined like degree, because application environment is similar, can make it that model to be measured is more close, so comparing the similarity of application environment, just
More like model to be measured can be found.
Third embodiment of the present invention is related to a kind of adaptive acquiring method of water quality soft-sensing model.3rd embodiment party
Formula is roughly the same with first embodiment, is in place of the main distinction:In third embodiment of the invention, the first hard measurement mould
Type is sampled in the water environment that model to be selected is surveyed, and the first soft-sensing model is tested using institute's water sampling, to sampling sample
This is tested, and can avoid carrying out On-line sampling system, and uses off-line measurement mode instead, avoids using on-line measurement institute as far as possible
The equipment that need to be used, further reduces testing cost.
In testing the first soft-sensing model, specifically include:Sampled in the water environment that model to be selected is surveyed;Utilize
Institute's water sampling is tested the first soft-sensing model., only need to be in model to be selected when testing the first soft-sensing model
Sample, the water sample after sampling, can measure in any place, any time in the water environment of survey, surveyed in real time without online
Data are measured, but have carried out offline measurement, therefore, just reduce the equipment used during On-line sampling system data, so as to
To play the effect for reducing testing cost.
In terms of existing technologies, the main distinction and effect are present embodiment:First soft-sensing model is to be selected
Sampled in the water environment that model is surveyed, the first soft-sensing model is tested using institute's water sampling, sample is surveyed
Examination, it can avoid carrying out On-line sampling system, and use off-line measurement mode instead, avoid as far as possible using being used needed for on-line measurement
Equipment, further reduce testing cost.
The 4th embodiment of the present invention is related to a kind of adaptive acquiring method of water quality soft-sensing model.4th embodiment party
Formula is roughly the same with first embodiment, is in place of the main distinction:In four embodiment of the invention, if matched
The first soft-sensing model quantity be more than 1, then according to test result accuracy of selection the first soft-sensing model of highest, and
Adjust the parameter of first soft-sensing model so that allow the first soft-sensing model quantity more than one matched, so as to
Therefrom to select a soft-sensing model of degree of accuracy highest first, and its parameter is adjusted.
A kind of flow chart of the adaptive acquiring method of the water quality soft-sensing model of this 4th embodiment, as shown in Figure 2.
Because step 201 to 202 and step 101 are to 102 completely the same, and step 206 to 208 with step 104 to 106 complete
Complete consistent, this is no longer going to repeat them.
Step 203:From water quality soft-sensing model storehouse, the first soft-sensing model is matched.
Specifically, present embodiment can select the similarity of characteristic information to be more than all first soft of preset percentage
Measurement model, then the first soft-sensing model quantity that step 203 matches are likely larger than 1.
Step 204:Whether the first soft-sensing model quantity for judging to match is more than 1.
Specifically, if the first soft-sensing model quantity matched is more than 1, into step 205, otherwise,
Into in step 206.The the first soft-sensing model quantity matched can be multiple, when quantity is more than 1, can there is more multiselect
The chance selected, now, into step 205, further operated.
It is noted that can also be when matching in present embodiment, restriction matches the similar of characteristic information
5 soft-sensing models are as the first soft-sensing model before spending highest, then, it will also match multiple first soft-sensing models.
, can also be according to 6 or 7 before the similarity highest of characteristic information it should be noted that in actual applications
One soft-sensing model is selected, wherein, the quantity of similarity highest model, it can be set by user.In addition, selection is special
The similarity of reference breath is more than in all first soft-sensing models of preset percentage, and preset percentage can also be carried out by user
It is determined that such as 95%, 98% etc..The value of user's setting is different, then the qualified first soft-sensing model quantity matched
For more than or equal to 1.
Step 205:Obtain the degree of accuracy the first soft-sensing model of highest.
Specifically, if the quantity of the first soft-sensing model matched is more than 1, selected according to test result
The soft-sensing model of degree of accuracy highest first.Using the soft-sensing model of degree of accuracy highest first as the soft survey further adjusted
Model is measured, continues step 206.
In terms of existing technologies, the main distinction and effect are present embodiment:If first matched is soft
The quantity of measurement model is more than 1, then according to test result accuracy of selection the first soft-sensing model of highest, and adjust this
The parameter of one soft-sensing model so that allow the first soft-sensing model quantity more than one matched, so as to therefrom select
A soft-sensing model of degree of accuracy highest first is selected, and its parameter is adjusted.
The step of various methods divide above, be intended merely to describe it is clear, can be merged into when realizing a step or
Some steps are split, are decomposed into multiple steps, as long as including identical logical relation, all protection domain in this patent
It is interior;To either adding inessential modification in algorithm in flow or introducing inessential design, but its algorithm is not changed
Core design with flow is all in the protection domain of the patent.
Fifth embodiment of the invention is related to a kind of adaptive acquisition device of water quality soft-sensing model, as shown in figure 3, bag
Include:
Memory module 31, for prestoring, multiple water quality soft-sensing models form water quality soft-sensing model storehouse, are that each water quality is soft
Measurement model adds characteristic information.
Communication module 32, for receiving the characteristic information of model to be selected.
Processing module 33, for utilizing received characteristic information, from water quality soft-sensing model storehouse, according to received
Characteristic information and each water quality soft-sensing model characteristic information similarity, match the first soft-sensing model.
Test module 34, for testing the first soft-sensing model, it is no more than preset value in the degree of accuracy tested
When, the parameter of the first soft-sensing model is adjusted, until the degree of accuracy tested reaches preset value.
Output module 35, for the degree of accuracy tested to be reached to the first soft-sensing model of preset value as adaptively obtaining
The water quality soft-sensing model got.
Memory module 31, it is additionally operable to the water quality soft-sensing model deposit water quality soft-sensing model storehouse that will adaptively get.
In terms of existing technologies, the main distinction and effect are present embodiment:Utilize existing water quality hard measurement mould
The similarity mode of type and the water quality soft-sensing model received, the first closest soft-sensing model is found out, and using not
Disconnected test, the accommodation of parameter is localized, so as to which quick obtaining meets the soft-sensing model of water quality to be measured, greatly
The formation speed of water quality soft-sensing model is accelerated, new water quality soft-sensing model is regenerated so as to avoid, reduces mould
The manufacturing cost of type.
It is seen that present embodiment is the system embodiment corresponding with first embodiment, present embodiment can be with
First embodiment is worked in coordination implementation.The relevant technical details mentioned in first embodiment still have in the present embodiment
Effect, in order to reduce repetition, is repeated no more here.Correspondingly, the relevant technical details mentioned in present embodiment are also applicable in
In first embodiment.
It is noted that each module involved in present embodiment is logic module, and in actual applications, one
Individual logic unit can be a part for a physical location or a physical location, can also be with multiple physics lists
The combination of member is realized.In addition, in order to protrude the innovative part of the present invention, will not be with solving institute of the present invention in present embodiment
The unit that the technical problem relation of proposition is less close introduces, but this is not intended that in present embodiment and other lists are not present
Member.
Sixth embodiment of the invention is related to a kind of adaptive acquisition device of water quality soft-sensing model.6th embodiment
It is roughly the same with the 5th embodiment, it is in place of the main distinction:In sixth embodiment of the invention, characteristic information is according to water quality
The application environment setting of soft-sensing model, because application environment is similar, can make it that model to be measured is more close, so comparing application
The similarity of environment, it is possible to find more like model to be measured.
Characteristic information is set according to the application environment of water quality soft-sensing model;Received characteristic information and each water quality are soft
The similarity of the characteristic information of measurement model, it is true according to the similarity of model to be selected and the application environment of water quality soft-sensing model
It is fixed.
In terms of existing technologies, the main distinction and effect are present embodiment:Characteristic information is according to the soft survey of water quality
The application environment setting of model is measured, because application environment is similar, can make it that model to be measured is more close, so comparing application environment
Similarity, it is possible to find more like model to be measured.
Because second embodiment is mutually corresponding with present embodiment, therefore present embodiment can be mutual with second embodiment
It is engaged implementation.The relevant technical details mentioned in second embodiment are still effective in the present embodiment, implement second
The technique effect that can reach in mode can similarly be realized in the present embodiment, no longer superfluous here in order to reduce repetition
State.Correspondingly, the relevant technical details mentioned in present embodiment are also applicable in second embodiment.
It will be appreciated by those skilled in the art that realize that all or part of step in above-described embodiment method is to pass through
Program instructs the hardware of correlation to complete, and the program storage is in the storage medium, including some instructions are causing one
Individual equipment (can be single-chip microcomputer, chip etc.) or processor (processor) perform each embodiment methods described of the application
All or part of step.And foregoing storage medium includes:USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only
Memory), random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various can store journey
The medium of sequence code.
It will be understood by those skilled in the art that the respective embodiments described above are to realize the specific embodiment of the present invention,
And in actual applications, can to it, various changes can be made in the form and details, without departing from the spirit and scope of the present invention.
Claims (10)
- A kind of 1. adaptive acquiring method of water quality soft-sensing model, it is characterised in that including:Multiple water quality soft-sensing models that prestore form water quality soft-sensing model storehouse, and feature is added for each water quality soft-sensing model Information;Receive the characteristic information of model to be selected;Received characteristic information is utilized, from the water quality soft-sensing model storehouse, according to received characteristic information and respectively The similarity of the characteristic information of the water quality soft-sensing model, matches the first soft-sensing model;First soft-sensing model is tested, when the degree of accuracy tested is no more than preset value, adjustment described first The parameter of soft-sensing model, until the degree of accuracy tested reaches the preset value;The degree of accuracy tested is reached into the first soft-sensing model of the preset value as the water quality adaptively got Soft-sensing model.
- 2. the adaptive acquiring method of water quality soft-sensing model according to claim 1, it is characterised in that also include:The water quality soft-sensing model adaptively got is stored in the water quality soft-sensing model storehouse.
- 3. the adaptive acquiring method of water quality soft-sensing model according to claim 1, it is characterised in that the feature letter Breath is set according to the application environment of the water quality soft-sensing model;The similarity of the characteristic information of the received characteristic information and each water quality soft-sensing model, modeling is treated according to described The similarity of the application environment of type and the water quality soft-sensing model determines.
- 4. the adaptive acquiring method of water quality soft-sensing model according to claim 3, it is characterised in that described to apply ring Border includes one below or its any combination:Water treatment technology used by water quality index to be measured, water to be measured, geographical position residing for water to be measured.
- 5. the adaptive acquiring method of water quality soft-sensing model according to claim 4, it is characterised in that if described should Included with environment:Water treatment technology used by the water to be measured, then according to technique correlation, determine that the water to be measured is used Water treatment technology similarity.
- 6. the adaptive acquiring method of water quality soft-sensing model according to claim 1, it is characterised in that described to first During soft-sensing model is tested, specifically include:Sampled in the water environment that the model to be selected is surveyed;First soft-sensing model is tested using institute's water sampling.
- 7. the adaptive acquiring method of water quality soft-sensing model according to claim 1, it is characterised in that if matched The quantity of the first soft-sensing model gone out is more than 1, then described that first soft-sensing model is tested, and is adjusted according to test result In the step of parameter of whole first soft-sensing model, specifically include:According to test result accuracy of selection the first soft-sensing model of highest, and adjust the parameter of first soft-sensing model.
- A kind of 8. adaptive acquisition device of water quality soft-sensing model, it is characterised in that including:Memory module, for prestoring, multiple water quality soft-sensing models form water quality soft-sensing model storehouse, are each soft survey of water quality Measure model addition characteristic information;Communication module, for receiving the characteristic information of model to be selected;Processing module, for utilizing received characteristic information, from the water quality soft-sensing model storehouse, according to received The similarity of the characteristic information of characteristic information and each water quality soft-sensing model, matches the first soft-sensing model;Test module, for testing first soft-sensing model, when the degree of accuracy tested is no more than preset value, The parameter of first soft-sensing model is adjusted, until the degree of accuracy tested reaches the preset value;Output module, for the degree of accuracy tested to be reached to the first soft-sensing model of the preset value as adaptive The water quality soft-sensing model got.
- 9. the adaptive acquisition device of water quality soft-sensing model according to claim 8, it is characterised in that the storage mould Block, it is additionally operable to the water quality soft-sensing model adaptively got being stored in the water quality soft-sensing model storehouse.
- 10. the adaptive acquisition device of water quality soft-sensing model according to claim 8, it is characterised in that the feature Information is set according to the application environment of the water quality soft-sensing model;The similarity of the characteristic information of the received characteristic information and each water quality soft-sensing model, modeling is treated according to described The similarity of the application environment of type and the water quality soft-sensing model determines.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610601948.1A CN107664680B (en) | 2016-07-27 | 2016-07-27 | Self-adaptive acquisition method and device of water quality soft measurement model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610601948.1A CN107664680B (en) | 2016-07-27 | 2016-07-27 | Self-adaptive acquisition method and device of water quality soft measurement model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107664680A true CN107664680A (en) | 2018-02-06 |
CN107664680B CN107664680B (en) | 2020-08-14 |
Family
ID=61115291
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610601948.1A Active CN107664680B (en) | 2016-07-27 | 2016-07-27 | Self-adaptive acquisition method and device of water quality soft measurement model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107664680B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109187534A (en) * | 2018-08-01 | 2019-01-11 | 江苏凯纳水处理技术有限公司 | Water quality detection method and its water sample pattern recognition device |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110295777A1 (en) * | 2010-06-01 | 2011-12-01 | National Tsing Hua University | Method for building adaptive soft sensor |
CN103295165A (en) * | 2013-06-26 | 2013-09-11 | 上海铂尔怡环境技术股份有限公司 | Water quality environmental quality assessment pre-warning system and method |
CN103810309A (en) * | 2012-11-08 | 2014-05-21 | 连晓峰 | Soft measurement modeling method of A2O municipal sewage treatment process based on constraint theory |
CN104318325A (en) * | 2014-10-14 | 2015-01-28 | 广东省环境监测中心 | Multi-basin real-time intelligent water quality predication method and system |
CN105205224A (en) * | 2015-08-28 | 2015-12-30 | 江南大学 | Modeling method for soft measurement of time difference gaussian process regression based on fuzzy curve analysis |
CN105630656A (en) * | 2014-11-06 | 2016-06-01 | 阿里巴巴集团控股有限公司 | Log model based system robustness analysis method and apparatus |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
SG10201406850VA (en) * | 2014-10-23 | 2016-05-30 | Sembcorp Ind Ltd | System and method for predicting a parameter of wastewater treatment process |
-
2016
- 2016-07-27 CN CN201610601948.1A patent/CN107664680B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110295777A1 (en) * | 2010-06-01 | 2011-12-01 | National Tsing Hua University | Method for building adaptive soft sensor |
CN103810309A (en) * | 2012-11-08 | 2014-05-21 | 连晓峰 | Soft measurement modeling method of A2O municipal sewage treatment process based on constraint theory |
CN103295165A (en) * | 2013-06-26 | 2013-09-11 | 上海铂尔怡环境技术股份有限公司 | Water quality environmental quality assessment pre-warning system and method |
CN104318325A (en) * | 2014-10-14 | 2015-01-28 | 广东省环境监测中心 | Multi-basin real-time intelligent water quality predication method and system |
CN105630656A (en) * | 2014-11-06 | 2016-06-01 | 阿里巴巴集团控股有限公司 | Log model based system robustness analysis method and apparatus |
CN105205224A (en) * | 2015-08-28 | 2015-12-30 | 江南大学 | Modeling method for soft measurement of time difference gaussian process regression based on fuzzy curve analysis |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109187534A (en) * | 2018-08-01 | 2019-01-11 | 江苏凯纳水处理技术有限公司 | Water quality detection method and its water sample pattern recognition device |
Also Published As
Publication number | Publication date |
---|---|
CN107664680B (en) | 2020-08-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Waldherr et al. | Violation of a temporal Bell inequality for single spins in a diamond defect center | |
CN110519128B (en) | Random forest based operating system identification method | |
Xu et al. | Feasibility study on the least square method for fitting non-Gaussian noise data | |
CN107968987B (en) | RSSI weighted centroid positioning method based on fixed integral combined with environmental parameters | |
CN111366820A (en) | Pattern recognition method, device, equipment and storage medium for partial discharge signal | |
CN106556877B (en) | A kind of earth magnetism Tonghua method and device | |
CN108540267A (en) | A kind of multi-user data information detecting method and device based on deep learning | |
CN111898734A (en) | NMR (nuclear magnetic resonance) relaxation time inversion method based on MLP (Multi-layer linear programming) | |
Huang | Comparison of three approaches for computing measurement uncertainties | |
CN111723523A (en) | Estuary residual water level prediction method based on cascade neural network | |
CN107664680A (en) | A kind of adaptive acquiring method and device of water quality soft-sensing model | |
CN110619216A (en) | Malicious software detection method and system for adversarial network | |
Barbini et al. | Weak signal detection based on two dimensional stochastic resonance | |
CN113726545A (en) | Network traffic generation method and device for generating countermeasure network based on knowledge enhancement | |
CN106662463A (en) | Method for detecting noise of sensor background and device | |
CN116306268A (en) | Shield tunneling simulation model parameter identification method system based on federal reinforcement learning | |
CN114064459A (en) | Software defect prediction method based on generation countermeasure network and ensemble learning | |
CN112532562B (en) | Malicious data flow detection method and system for adversarial network | |
CN107067036A (en) | A kind of ground net corrosion rate prediction method | |
CN105224569B (en) | A kind of data filtering, the method and device for constructing data filter | |
CN104200510B (en) | Vector quantization compression object plotting method based on target CF | |
CN117670139B (en) | Intelligent detection method and system for PP line quality | |
Wang et al. | A new simulation algorithm for absorbing receiver in molecular communication | |
CN114440888B (en) | Indoor positioning method and device based on sequence grouping sliding window | |
Kim | Comparison of structural change tests in linear regression models |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |