CN109273058A - A kind of composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid - Google Patents
A kind of composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid Download PDFInfo
- Publication number
- CN109273058A CN109273058A CN201811098049.XA CN201811098049A CN109273058A CN 109273058 A CN109273058 A CN 109273058A CN 201811098049 A CN201811098049 A CN 201811098049A CN 109273058 A CN109273058 A CN 109273058A
- Authority
- CN
- China
- Prior art keywords
- early warning
- data
- fatty acid
- volatile fatty
- value
- 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
Landscapes
- Purification Treatments By Anaerobic Or Anaerobic And Aerobic Bacteria Or Animals (AREA)
Abstract
The invention proposes a kind of composite algorisms for the exceeded early warning of anaerobic processes volatile fatty acid, comprising: collecting sample data set establishes decision table；According to the sample data set in decision table, principal component analysis is carried out, obtains principal component table and initial Factor load-matrix table；According to principal component table and initial Factor load-matrix table, feature vector is calculated, and calculates the accounting for obtaining each index in overall data；The index value that accounting is high in accounting data is chosen, by the breakpoint discretization in the corresponding data set of index value, breakpoint is chosen and forms corresponding section；It generates initial rules collection and the final rule set of reactor is generated according to initial rules collection and the section；Using the data for participating in rough set excavation as test set, test assessment is carried out using the final rule set, obtains three decision value range correct classification rates of reactor, passed through rough set and carry out the exceeded early warning of volatile fatty acid.
Description
Technical field
It is espespecially a kind of for the exceeded early warning of anaerobic processes volatile fatty acid the present invention relates to anaerobism early warning technology field
Composite algorism.
Background technique
Anaerobic digestion process is one and extremely complex is related to a large amount of, multi-step microbiological inside microbial cell
In chemical reaction process and anaerobic reactor the alternate macroscopic quality of solid-liquid-gas three transmitting, heat transfer and energy transmission mistake
Journey.Fermentation process includes complicated parameter, and is highly coupled, influences each other between each parameter.Control is monitored to fermentation process
System, may advantageously facilitate fermentation efficiency and improves the stability and safety of anaerobic fermentation process.Due to existing in anaerobic reactor
A variety of buffer systems make indicators for the buffer function of pH value --- volatile fatty acid (VFAs,volatile
Fatty acids) accumulation to a certain extent cannot be reflected in time on pH, therefore directly measured and controlled VFAs and refer to
Scale value becomes the effective ways of anaerobic reactor operation control, VFA in the reasonable scopesDetection method and its detection efficiency
As the key that can judge anaerobic technique operating status in time.
In anaerobic reaction system, acidification bacteria and methanogen are in commensalism, and methanogen utilizes acidification bacteria
Volatile fatty acid (the VFA that decomposing organic matter generatess, volatile fatty acids) and it is used as metabolin, it decomposes and generates first
Substances such as alkane, carbon dioxide, while generating basicity, in the anaerobic system of operational excellence, organic acid that acidification bacteria generates and
The basicity that methanogen generates is in the state of relative equilibrium, this makes the pH value of anaerobic digestion solution maintain stable range
It is interior, acid or alkali environment needed for meeting microorganism eubolism.But methane backeria ought be caused to volatile fatty acid for some reason
Utilization rate reduce when (such as temperature fluctuation, the influence of load impact, toxicant) will result in VFAsAccumulation, when this product
When tiring out lasting carry out to a certain extent, acid-base balance is just easy to be broken, and it is " sour that the pH of fermentation liquid can decline suddenly at this time causes
Lose " appearance of phenomenon, once appearance is rancid, the methanogen population activity in reactor will be heavily suppressed, rebuild acidification bacterium and
The ecological balance of methane backeria will take a substantial amount of time and put into a large amount of drug, and take empties anaerobic reaction more in Practical Project
Device, the measure that renewed vaccination sludge is restarted, this will affect normal sewage treatment process, even result in entire sewage treatment system
The paralysis of system makes enterprise suffer huge economic losses and environmental management risk.Therefore, VFA is continuously monitoredsIndex also just becomes crowd
Important one work in more anaerobic system operational process.And to the detection of VFAs be merely capable of characterization system operation history or
Accomplished fact, and anaerobism operational process is a complicated biochemical reaction process, and the variation of VFAs is passed through by multifactor impact
Whether the relationship instruction VFAs between index can be used as a kind of necessity more than secure threshold in the level being contemplated that in the period
Early warning mechanism to guarantee the stabilization and safety of anaerobic reaction process.
Traditional includes the way of distillation, colorimetric method, gas chromatography, titration using chemical analysis or Instrument measuring VFAs
Deng.In addition, in a manner of hard measurement by neural network, fuzzy control theory, support vector machines etc. for all kinds of mathematical algorithms of representative
Also become a branch in this field, domestic and international some scholars did research work to the hard measurement of the parameter.
Traditional chemical analysis method such as the way of distillation, titration needs manual sampling to analyze, and analysis frequency is low, by chemical examination point
The influence of analysis personnel operation, analyzes the repeatability of result, reproducibility cannot be guaranteed；Gas chromatography, colorimetric method etc. pass through instrument
Method for measuring the degree of automation is low, and needing artificial sample and do can examination with computer after the pretreatment of sample.Above-mentioned materialization
Method is to VFAsThe test only process that has occurred that in reflection reactor, cannot reflect that the following a cycle index may
Level.
Flexible measurement method with neural network, fuzzy theory, support vector machines etc. for representative is with each of reactor operation
Index also includes VFA such as temperature, pH, loadsEtc. parameters be input pointer, with VFAsFor output-index, founding mathematical models,
Using above-mentioned each metric history operation data as sample training mathematical model and target component is predicted, to realize the soft survey of parameter
Amount, its advantage is that next period VFA can be predicted according to current data informationsLevel, to realize prediction and early warning function
Energy.The shortcomings that above method, is: 1, being easy to be influenced by shortage of data during practical engineering application；2, when data acquire frequency
When rate is inconsistent, the frequency for such as sampling analysis data is mostly 1~3 times/day, and the frequency of sensor online data is mostly several
Second or a few minutes are primary, this just needs to do data the synchronous pretreatment of frequency, are readily incorporated invalid data or loss details, lead
Cause the failure of prediction.
Obviously, under the conditions of data volume is sufficient and information is accurate, flexible measurement method can pass through founding mathematical models
Realization is predicted data infinite approach detected value, and in relevant research report, this prediction mode has also passed through all kinds of experiments
Verifying, but it is dry that imperfect data information, shortage of data, dirty data, signal often occurs in engineering actual moving process
Equal various factors are disturbed, these factors will impact the calculating to mathematical model, will lead to prediction of failure when serious.
Summary of the invention
The present invention proposes a kind of VFA based on Rough Set TechniquesPrediction technique, rough set (Rough set) technology are places
Manage the effective means of the various incomplete information such as inaccurate, inconsistent, imperfect.Rough set theory is after probability theory, fuzzy set
Another probabilistic mathematical tool of processing theoretical, after evidence theory.It is coarse as a kind of newer hydropower unit
Collection is increasingly taken seriously in recent years, and validity is confirmed in the successful application in many scientific and engineering fields, is
Currently one of the research hotspot in artificial intelligence theory and its application field in the world；There is not yet Rough Set Technique is applied to anaerobism
Course prediction and early warning field.
Specifically, the algorithm includes: S1, collecting sample data set establishes decision table；S2, according to the sample in decision table
Data set carries out principal component analysis, obtains principal component table and initial Factor load-matrix table；S3, according to principal component table and initially
Factor load-matrix table calculates feature vector using formula (1), and calculates the accounting for obtaining each index in overall data；
S4 chooses the index value that accounting is high in accounting data, by the breakpoint discretization in the corresponding data set of index value, chooses breakpoint simultaneously
Form corresponding section；S5 generates initial rules collection according to initial rules collection and the section and generates the final rule of reactor
Then collect；S6 carries out test assessment using the final rule set using the data for participating in rough set excavation as test set, obtains
Three decision value range correct classification rates of reactor carry out the exceeded early warning of volatile fatty acid by rough set.
The above-mentioned composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid has following technical effect that
1, rule is established using Rough Set Technique to realize to anaerobic digestion process key parameter VFAsThe prediction of distributed area,
And early warning is carried out to transfiniting；
2, the classification of input pointer and quantity are not required, is suitble to the index prediction of not priori knowledge；
3, invalid information is removed by index dimension-reduction treatment, improves computational efficiency, reduces the complexity of Rule.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, not
Constitute limitation of the invention.In the accompanying drawings:
Fig. 1 is the step flow diagram of one embodiment of the invention.
Specific embodiment
Cooperate schema and presently preferred embodiments of the present invention below, the present invention is further explained to reach predetermined goal of the invention institute
The technological means taken.
As shown in Figure 1, in conjunction with step S1, it is necessary first to establish decision table.Decision table is mainly by object, attribute and attribute value
Composition.Object refers to that the sample data set acquired by date, attribute are the anaerobic processes parameter that can be obtained, and attribute value refers to each ginseng
Several data.In order to play conditional attribute as much as possible to the influence of decision attribute, also for making full use of available data,
By inflow (F), influent COD, HLR, ALR, inlet flow-patterm, inflow temperature, TSS, water inlet VFA, reactor pH, temperature in tank, most
High temperature, the lowest temperature and when temperature difference per day is used as conditional attribute, each conditional attribute and its code name are defined and are shown in Table 1.
1 conditional attribute of table and its code name definition
In conjunction with step S2, principal component analysis (Principal Component Analysis, PCA) is to investigate multiple variables
Between correlation a kind of multivariate statistical method, by orthogonal transformation, there may be the variables of correlation to be converted to one group of line by one group
The incoherent variable of property, this group of variable after conversion is principal component.The core concept of Principal Component Analysis is dimensionality reduction, this method
Multi objective can be subjected to optimum comprehensive simplification, be eventually converted into less overall target.Therefore, using Principal Component Analysis,
The workload of data statistics in evaluation procedure can be simplified, eliminate the information overlap between index.
1, original index data as input are as shown in table 1.
2, it is analyzed using SPSS software principal component, obtains Principle component extraction table (as shown in table 2) and the initial factor carries
Lotus matrix table (as shown in table 3).
2 Principle component extraction table of table
The initial factor loading table of table 3
Feature vector, i.e. principal component are calculated using initial Factor load-matrix and principal component characteristic value in conjunction with step S3
Expression formula coefficient A, its calculation formula is: the initial load factor/SQR (corresponding principal component characteristic value).Later, each index is calculated
The shared ratio in entire data, the results are shown in Table 4.
Each achievement data ratio table of table 4
Same principal component analysis is done according to the data that above-mentioned steps do reactor to other two, obtains the ratio of each index
Example result.The principal component analysis of three groups of samples the results are shown in Table 5.
Each achievement data ratio table in 5 three groups of samples of table
In conjunction with step S4, by principal component analysis it is found that the higher index of accounting include: inflow temperature, temperature in tank,
HLR, the local highest temperature and the local lowest temperature, illustrate that these indexs are higher than other to the influence degree of anaerobism operational process
Index.And it is smaller to work as ratio shared by the indexs such as temperature difference per day, therefore can ignore its influence to conceptual data, realizes original number
According to dimensionality reduction operation.
Although conditional attribute index value in initial data is discrete, but dispersion degree is inadequate.We are by Rosetta
Equal frequency binning algorithm (EFBA) algorithm provided by software carries out the breakpoint discretization fortune of data
It calculates (table 6).Conditional attribute value can be divided into several sections by breakpoint, be used to carry out section classification, class to input pointer data
It is similar to the section definition to VFA, we indicate different sections with the mode of number.Such as the inflow of 1#IC reactor, two
A breakpoint 6840 and 8520 forms three sections [- ∞, 6840], [6840,8520], [8520 ,+∞].In this fact Example, with
1,2,3 these three different sections of expression, other indexs and so on.
Break point set of the table 6 based on EFBA algorithm
In conjunction with step S5, by Rosetta Software Create initial rules collection.Choose initial rules concentrate accuracy be greater than etc.
In 0.45, relevance grade is more than or equal to 0.02 several rules.3 final rule sets of IC reactor are respectively as shown in table 7-9.
The 7 final rule set of 1#IC reactor of table
The 8 final rule set of 2#IC reactor of table
The 9 final rule set of 3#IC reactor of table
In conjunction with step S6, rules evaluation is carried out.
Again using the data for participating in rough set excavation as test set, test assessment, gained are carried out using above-mentioned rule set
As a result as shown in table 10-12.Rough set is 69.56% respectively to the correct classification rate of three reactors, three decision value ranges,
64.79% and 64.73%, although not high for 1 and 2 two kinds of situation accuracys rate to decision value, determine to we are of interest
Accuracy rate when plan value is 3 has then respectively reached 83.18%, 82.57% and 80.19%, and the regular quantity excavated is obvious
More than other two decision attribute values, show Rough Set Technique in terms of the prediction of VFA high range more precisely.
The rule set of 10 1#IC reactor of table is assessed
The rule set of 11 2#IC reactor of table is assessed
The rule set of 12 3#IC reactor of table is assessed
When using above-mentioned composite algorism, not only VFAsIt can be established and be predicted by this method, the prediction of other parameters is same
Sample can use this method.Breakpoint section is not limited to 3, can divide more smaller sections by introducing multiple breakpoints,
To keep the range of results of prediction more accurate.Priori knowledge, such as anaerobic digestion can be introduced for the selection of data sample breakpoint
The normal range (NR) of process pH value is 6.8~7.2.The composite algorism can realize the automation of prediction process by programming, improve
Computational efficiency.
In order to carry out apparent explanation to above-mentioned composite algorism, said below with reference to a specific embodiment
It is bright, the present invention is improperly limited however, it should be noted that the embodiment merely to the present invention is better described, is not constituted
It is fixed.
1, in daily data sample, there is 5 during on October 15,15 days to 2016 April in 2016 for 2#IC reactor tank
Months 31 days, June 1, September 15 days water outlet VFA value 120 or more.Breakpoint and its formation according to each index of correlation in table 7
Codomain, the value condition of each ATTRIBUTE INDEX on these three sample dates is as shown in table 13.
The index situation on table three sample dates of 13 2#IC reactor tank
The 2#IC reactor tank rule obtained by this project, (16) and (25) rule can be used to give for sample 1 and sample 2
Determine.The decision value of sample 1 and sample 2 is 3, that is, water outlet VFA value is 120 or more.
2, in daily data sample, there is 5 during on October 15,15 days to 2016 April in 2016 for 3#IC reactor tank
The water outlet VFA value on the moon 4 is 120 or more.According to the breakpoint for each index of correlation that this project determines, this is listed shown in table 14
The value condition of each ATTRIBUTE INDEX on three sample dates.
The index situation on 14 some sample date of 3#IC reactor tank of table
The above-mentioned composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid has following technical effect that
1, rule is established using Rough Set Technique to realize to anaerobic digestion process key parameter VFAsThe prediction of distributed area,
And early warning is carried out to transfiniting；
2, the classification of input pointer and quantity are not required, is suitble to the index prediction of not priori knowledge；
3, invalid information is removed by index dimension-reduction treatment, improves computational efficiency, reduces the complexity of Rule.
Particular embodiments described above has carried out further in detail the purpose of the present invention, technical scheme and beneficial effects
Describe in detail it is bright, it should be understood that the above is only a specific embodiment of the present invention, the guarantor being not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in this
Within the protection scope of invention.
Claims (6)
1. a kind of composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid characterized by comprising
S1, collecting sample data set, establishes decision table；
S2 carries out principal component analysis according to the sample data set in decision table, obtains principal component table and initial Factor load-matrix
Table；
S3 calculates feature vector using formula (1), and it is every to calculate acquisition according to principal component table and initial Factor load-matrix table
Accounting of a index in overall data；
Feature vector=initial load the factor/corresponding principal component characteristic value；Formula (1)
S4 chooses the index value that accounting is high in accounting data, and the breakpoint discretization in the corresponding data set of index value is chosen disconnected
It puts and forms corresponding section；
S5 generates initial rules collection according to initial rules collection and the section and generates the final rule set of reactor；
S6 carries out test assessment using the final rule set using the data for participating in rough set excavation as test set, obtains anti-
Three decision value range correct classification rates for answering device carry out the exceeded early warning of volatile fatty acid by rough set.
2. the composite algorism according to claim 1 for the exceeded early warning of anaerobic processes volatile fatty acid, feature exist
In, in step sl, decision table is mainly made of object, attribute and attribute value, and object is the sample data set acquired by date,
Attribute is the anaerobic processes parameter that can be obtained, and attribute value refers to the data of each parameter；Wherein, attribute includes: inflow, water inlet
COD, HLR, ALR, inlet flow-patterm, inflow temperature, TSS, water inlet VFA, reactor pH, temperature in tank, the highest temperature, the lowest temperature and
Work as temperature difference per day.
3. the composite algorism according to claim 2 for the exceeded early warning of anaerobic processes volatile fatty acid, feature exist
In, in step s 2, utilize SPSS software carry out principal component analysis.
4. the composite algorism according to claim 3 for the exceeded early warning of anaerobic processes volatile fatty acid, feature exist
In in step s 4, choosing accounting in accounting data is positive value and the index value high to anaerobism operational process influence degree.
5. the composite algorism according to claim 4 for the exceeded early warning of anaerobic processes volatile fatty acid, feature exist
In, in step s 4, introducing two values [b1,b2] be used as breakpoint, make data sample be divided to be formed three sections [- ∞,
b1]、[b1, b2]、[b2,+∞], classification of these three sections as decision attribute values；
As digestive juice VFAsIn section [- ∞, b1] when, d value is 1；As digestive juice VFAsIn section [b1, b2] when, d value
It is 2；As digestive juice VFAsIn section [b2,+∞] when, d value is 3；As digestive juice VFAsPrediction is in section [b2,+∞]
When, indicate VFAsIn higher level, need to make exceeded early warning.
6. the composite algorism according to claim 4 for the exceeded early warning of anaerobic processes volatile fatty acid, feature exist
In, in step s 5, according to Rosetta Software Create initial rules collection, chooses initial rules and concentrate accuracy >=0.45, it is applicable
The rule of degree >=0.02, generates the final rule set of reactor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811098049.XA CN109273058A (en) | 2018-09-20 | 2018-09-20 | A kind of composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811098049.XA CN109273058A (en) | 2018-09-20 | 2018-09-20 | A kind of composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109273058A true CN109273058A (en) | 2019-01-25 |
Family
ID=65197111
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811098049.XA Pending CN109273058A (en) | 2018-09-20 | 2018-09-20 | A kind of composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109273058A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112420161A (en) * | 2020-10-28 | 2021-02-26 | 西北农林科技大学 | Rumen volatile fatty acid prediction method, system, equipment and medium |
| CN114354867A (en) * | 2021-12-27 | 2022-04-15 | 江苏海峡环保科技发展有限公司 | Method for predicting volatile fatty acid in sewage |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853334A (en) * | 2010-05-27 | 2010-10-06 | 重庆大学 | Equation-free multi-scale simulation method in chemical reaction diffusion |
| CN102708677A (en) * | 2012-05-30 | 2012-10-03 | 杭州师范大学 | Traffic state evaluation rule extracting method based on rough set theory |
| CN103235098A (en) * | 2013-05-13 | 2013-08-07 | 安徽工程大学 | On-line ground water quality monitoring system |
| CN103454390A (en) * | 2013-03-27 | 2013-12-18 | 深圳信息职业技术学院 | Method and device for measuring concentration of dissolved oxygen |
| CN105045951A (en) * | 2015-05-27 | 2015-11-11 | 华南理工大学 | Soft-measurement method for volatile fatty acid in effluent of anaerobic wastewater treatment system |
| CN106600063A (en) * | 2016-12-16 | 2017-04-26 | 东北电力大学 | Space load prediction method based on fuzzy-rough set theory and spatio-temporal information |
| CN107016416A (en) * | 2017-04-12 | 2017-08-04 | 中国科学院重庆绿色智能技术研究院 | The data classification Forecasting Methodology merged based on neighborhood rough set and PCA |
| CN107132325A (en) * | 2017-04-14 | 2017-09-05 | 华南理工大学 | A kind of flexible measurement method of the Anaerobic Waste Treatment System water outlet volatile fatty acid based on particle cluster algorithm and SVMs |
| KR20180002280A (en) * | 2016-06-29 | 2018-01-08 | 경북대학교 산학협력단 | Computational system and method to identify cancer causing genes, recording medium for performing the method |
| CN107844679A (en) * | 2017-11-08 | 2018-03-27 | 中南大学 | A kind of segmented blast furnace molten iron silicon content Forecasting Methodology and device |
-
2018
- 2018-09-20 CN CN201811098049.XA patent/CN109273058A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853334A (en) * | 2010-05-27 | 2010-10-06 | 重庆大学 | Equation-free multi-scale simulation method in chemical reaction diffusion |
| CN102708677A (en) * | 2012-05-30 | 2012-10-03 | 杭州师范大学 | Traffic state evaluation rule extracting method based on rough set theory |
| CN103454390A (en) * | 2013-03-27 | 2013-12-18 | 深圳信息职业技术学院 | Method and device for measuring concentration of dissolved oxygen |
| CN103235098A (en) * | 2013-05-13 | 2013-08-07 | 安徽工程大学 | On-line ground water quality monitoring system |
| CN105045951A (en) * | 2015-05-27 | 2015-11-11 | 华南理工大学 | Soft-measurement method for volatile fatty acid in effluent of anaerobic wastewater treatment system |
| KR20180002280A (en) * | 2016-06-29 | 2018-01-08 | 경북대학교 산학협력단 | Computational system and method to identify cancer causing genes, recording medium for performing the method |
| CN106600063A (en) * | 2016-12-16 | 2017-04-26 | 东北电力大学 | Space load prediction method based on fuzzy-rough set theory and spatio-temporal information |
| CN107016416A (en) * | 2017-04-12 | 2017-08-04 | 中国科学院重庆绿色智能技术研究院 | The data classification Forecasting Methodology merged based on neighborhood rough set and PCA |
| CN107132325A (en) * | 2017-04-14 | 2017-09-05 | 华南理工大学 | A kind of flexible measurement method of the Anaerobic Waste Treatment System water outlet volatile fatty acid based on particle cluster algorithm and SVMs |
| CN107844679A (en) * | 2017-11-08 | 2018-03-27 | 中南大学 | A kind of segmented blast furnace molten iron silicon content Forecasting Methodology and device |
Non-Patent Citations (4)
| Title |
|---|
| 刘博 等，: "基于PCA-LSSVM的厌氧废水处理***出水VFA在线预测模型", 《环境科学学报》 * |
| 刘菁菁 等，: "基于主成分一粗糙集理论的企业信贷风险预警研究", 《哈尔滨商业大学学报（社会科学版）》 * |
| 张睿，: "数据分析在污染控制领域的节能优化应用", 《中国博士学位论文全文数据库工程科技Ⅰ辑》 * |
| 袁序成，: "《广西保险业发展与监管研究》", 30 September 2009, 广西民族出版社 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112420161A (en) * | 2020-10-28 | 2021-02-26 | 西北农林科技大学 | Rumen volatile fatty acid prediction method, system, equipment and medium |
| CN114354867A (en) * | 2021-12-27 | 2022-04-15 | 江苏海峡环保科技发展有限公司 | Method for predicting volatile fatty acid in sewage |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Regmi et al. | The future of WRRF modelling–outlook and challenges | |
| CN110320335B (en) | Multi-element robust soft measurement method for effluent quality index of sewage treatment | |
| Li et al. | Ensemble model of wastewater treatment plant based on rich diversity of principal component determining by genetic algorithm for status monitoring | |
| CN107132325B (en) | A kind of flexible measurement method based on the Anaerobic Waste Treatment System of particle swarm algorithm and support vector machines water outlet volatile fatty acid | |
| Liu et al. | Online prediction of effluent COD in the anaerobic wastewater treatment system based on PCA-LSSVM algorithm | |
| CN112417765B (en) | Sewage treatment process fault detection method based on improved teacher-student network model | |
| CN110232256A (en) | A kind of sewage disposal process monitoring method based on KPLS and RWFCM | |
| CN108764305A (en) | A kind of improved colony intelligence machine learning fault diagnosis system | |
| CN109273058A (en) | A kind of composite algorism for the exceeded early warning of anaerobic processes volatile fatty acid | |
| CN102778548B (en) | Method for forecasting sludge volume index in sewage treatment process | |
| CN115561416A (en) | Method and device for detecting inlet water quality of sewage treatment plant in real time | |
| CN117170221A (en) | Artificial intelligence control system for sewage treatment | |
| CN106706491B (en) | Intelligent detection method for membrane bioreactor-MBR water permeability | |
| CN108549740A (en) | A kind of anaerobic system water outlet ammonia nitrogen flexible measurement method based on integrated intelligent algorithm | |
| Cheng et al. | Knowledge and data-driven hybrid system for modeling fuzzy wastewater treatment process | |
| CN115294383A (en) | UMAP-GADF-CNN-based sewage treatment fault diagnosis method and system | |
| Yang et al. | Teacher–Student Uncertainty Autoencoder for the Process-Relevant and Quality-Relevant Fault Detection in the Industrial Process | |
| CN111204867B (en) | Membrane bioreactor-MBR membrane pollution intelligent decision-making method | |
| CN113702609A (en) | Water quality detection method and system | |
| CN109033524A (en) | A kind of chemical process concentration variable On-line Estimation method based on robust mixed model | |
| CN117113719A (en) | Sewage treatment simulation modeling system construction method | |
| Wu et al. | Process monitoring of nonlinear uncertain systems based on part interval stacked autoencoder and support vector data description | |
| Dixon et al. | Experience with data mining for the anaerobic wastewater treatment process | |
| Xie et al. | Fault diagnosis of large-scale industrial processes using the multi-block probabilistic kernel partial least squares method | |
| Komarysta et al. | OPTIMIZING BIOGAS PRODUCTION USING ARTIFICIAL NEURAL NETWORK. |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190125 |
|
| RJ01 | Rejection of invention patent application after publication |