CN108647783A - A kind of aquaculture water quality dissolved oxygen detection method - Google Patents
A kind of aquaculture water quality dissolved oxygen detection method Download PDFInfo
- Publication number
- CN108647783A CN108647783A CN201810440821.5A CN201810440821A CN108647783A CN 108647783 A CN108647783 A CN 108647783A CN 201810440821 A CN201810440821 A CN 201810440821A CN 108647783 A CN108647783 A CN 108647783A
- Authority
- CN
- China
- Prior art keywords
- dissolved oxygen
- data
- neural network
- value
- water quality
- 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.)
- Withdrawn
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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Pathology (AREA)
- Data Mining & Analysis (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Medicinal Chemistry (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Biochemistry (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Food Science & Technology (AREA)
- Farming Of Fish And Shellfish (AREA)
Abstract
The present invention provides a kind of aquaculture water quality dissolved oxygen detection method, belongs to aquaculture field.The data that this method acquires history are as data set, the artificial intelligence model of detection dissolved oxygen numerical value is established based on BP neural network, after neural metwork training is good, in the case where measuring dissolved oxygen without using dissolved oxygen sensor, dissolved oxygen numerical value is calculated at the time of acquisition according to temperature, turbidity, pH value and data.Because dissolving the expensive of oxygen detecting sensor, maintenance period is short, sundries absorption etc. can cause deviation of reading larger in water body, the present invention is according to the correlation and data fusion of multivariable, use neural network model, independent of dissolved oxygen sensor in the detection of practical dissolved oxygen, cost has been saved, while can solve the problems, such as dissolved oxygen read untrue caused by sundries absorption.
Description
Technical field
The invention belongs to aquaculture field, it is related to using neural network algorithm, especially usage history data training structure
The detection model of dissolved oxygen realizes the purpose of detection water quality dissolved oxygen without using dissolved oxygen sensor.
Background technology
Aquaculture is one of the important industry in China, and dissolved oxygen is to detect Cultivated water environment if appropriate for aquatic products
One of important parameter of cultivation.But detect oxygen in water concentration sensor it is expensive, be vulnerable to erosion damage or by
Dissolved oxygen sensor read untrue is caused to suspended matter absorption in water, a large amount of cost payout can be brought.In dissolved oxygen and water
Water plant and algae have prodigious relationship because water plant and algae are dissolved in water by the oxygen that photosynthesis generates
Amount is more than and comes from the amount that oxygen in air is dissolved in water, and photosynthesis with one at the time of, temperature, pH and turbidity it is related.
Therefore by measuring temperature, pH, turbidity and at the time of recording data acquisition, using multidata correlation and fusion, pass through
The method of artificial intelligence, you being implemented without dissolved oxygen sensor can achieve the purpose that dissolving allows detection.
Dissolving oxygen detecting sensor higher price currently on the market, and shorter in Cultivated water service life, easily by
The accuracy of data is easily influenced and reduced by adsorbate to erosion or dissolved oxygen sensor, so improving cultivation indirectly
The reliability of cost and data, therefore the present invention has important practical value.
Invention content
In order to reduce aquaculture cost, the present invention provides a kind of aquaculture water quality dissolved oxygen detection method.
Technical scheme of the present invention：
A kind of aquaculture water quality dissolved oxygen detection method, using the data of history acquisition as data set, based on BP nerves
Network establishes the artificial intelligence model of detection dissolved oxygen numerical value, after neural metwork training is good, without using dissolved oxygen sensor
In the case of measuring dissolved oxygen, dissolved oxygen numerical value is calculated at the time of acquisition according to temperature, turbidity, pH value and data, steps are as follows：
S1：At the time of dissolved oxygen numerical value, temperature, turbidity, pH value and the data of the Cultivated water of usage history acquisition acquire
Build data set；
S2：Cleaning and normalized are carried out to the data that data are concentrated, build sample set；
S3：Training set and test set are chosen from sample set in the method for cross validation, build BP neural network；With training
Temperature, turbidity, pH value and the data of concentration are trained as input variable, dissolved oxygen numerical value as output variable at the time of acquisition
BP neural network after BP neural network reaches required precision, establishes dissolving by the precision of test set test b P neural networks
Oxygen BP neural network model；
S4：At the time of acquiring the temperature of real-time Cultivated water, turbidity, pH value and record data acquisition, built with step S3
Vertical dissolved oxygen BP neural network model, is calculated the dissolved oxygen data of Cultivated water.
After the step S2 normalizeds, noise reduction is filtered to data first with median filtering method, then builds sample
This collection.
The cross validation of the step S3 is that sample set D is divided into k exclusive subsets Di by stratified sampling, each
Exclusive subsets Di keeps the consistency of data distribution；Then use the union of k-1 exclusive subsets as training set every time, it is remaining
Exclusive subsets form k group training/test sets as test set, k training of progress and test, and finally return that k test result
Mean value.
BP neural network in the step S3 contains two layers of hidden layer, when test set detection BP neural network reaches precision
After it is required that, the weights and threshold value of every layer of hidden layer are exported, dissolved oxygen BP neural network model is established.
Beneficial effects of the present invention：By the present invention in that with each change acquired in the detection process of historical data analysis waters
Correlation between amount finds that dissolved oxygen has prodigious correlation with temperature, time, turbidity and pH value.Then by using
The dissolved oxygen of historical data training structure calculates neural network model, and test result finds that BP neural network algorithm can be well
This correlation is fitted, to achieve the purpose that the data of temperature in use, time, turbidity and pH value calculate dissolved oxygen numerical value.
To saved in practical breeding process for detect dissolved oxygen sensor expense expenditure, solve dissolved oxygen sensor by
The inaccurate problem of the absorption reading of suspended material in water.
Description of the drawings
Fig. 1 is the design flow diagram of the present invention.
Fig. 2 shows the correlation of each variable and dissolved oxygen in the present invention.
Fig. 3 is BP network structure exemplary plots.
Fig. 4 is the design sketch of BP neural network prediction, and dotted line represents test data distribution, and solid line representative model predicts number
According to the two essentially coincides.
Specific implementation mode
The embodiment of the present invention is described in detail below in conjunction with technical solution and attached drawing.
Embodiment 1：A kind of aquaculture water quality dissolved oxygen detection method
S1：When using the dissolved oxygen numerical value of Cultivated water acquired in recent years, temperature, turbidity, pH value and corresponding acquisition
The data at quarter build data set, and it is column vector x to enable acquisition time1, temperature is column vector x2, turbidity is column vector x3, pH value is
x4, x1、x2、x3And x4Constitute matrix X0For data input set, dissolved oxygen Y is enabled0It is data output collection；
S2：The data acquired over the years are filtered and normalized, finally build sample set X and Y；
S3：Training set and test set are chosen with the method setting of cross validation；
BP neural network, BP network models example such as Fig. 3 is trained to be designed as using the training set of setting：
(1) input layer is：Xωin+bin=out1；
(2) hidden layer is：out1ωhi+bhi=out2；
(3) output layer is：
(4) adjuster library is used to adjust each threshold value and weights, training BP neural network；
It is finally assessed using test set test model precision and to model, uses following estimation items
If precision and assessment not up to require, re -training network or raising frequency of training；When BP neural network reaches
After required precision, dissolved oxygen BP neural network model is established；
S4：Data set is constituted at the time of acquisition real time temperature, turbidity, pH value and record data acquisition, with step S3 institutes
Dissolved oxygen data are calculated in the dissolved oxygen BP neural network model of foundation.
Claims (5)
1. a kind of aquaculture water quality dissolved oxygen detection method, which is characterized in that include the following steps：
S1：Dissolved oxygen numerical value, temperature, turbidity, pH value and the data of the Cultivated water of usage history acquisition are built at the time of acquisition
Data set；
S2：Cleaning and normalized are carried out to the data that data are concentrated, build sample set；
S3：Training set and test set are chosen from sample set in the method for cross validation, build BP neural network；With in training set
Temperature, turbidity, pH value and data be used as input variable, dissolved oxygen numerical value to train BP god as output variable at the time of acquire
Through network；By the precision of test set test b P neural networks, after BP neural network reaches required precision, dissolved oxygen BP is established
Neural network model；
S4：At the time of acquiring the temperature of real-time Cultivated water, turbidity, pH value and record data acquisition, established with step S3
The dissolved oxygen data of Cultivated water are calculated in dissolved oxygen BP neural network model.
2. a kind of aquaculture water quality dissolved oxygen detection method according to claim 1, which is characterized in that the step S2
After normalized, noise reduction is filtered to data first with median filtering method, then builds sample set.
3. a kind of aquaculture water quality dissolved oxygen detection method according to claim 1 or 2, which is characterized in that the step
The cross validation of rapid S3 is that sample set D is divided into k exclusive subsets Di, each exclusive subsets Di holdings by stratified sampling
The consistency of data distribution；Then use the union of k-1 exclusive subsets as training set every time, remaining exclusive subsets are as survey
Examination collection, forms k group training/test sets, carries out k training and test, finally returns that the mean value of k test result.
4. a kind of aquaculture water quality dissolved oxygen detection method according to claim 1 or 2, which is characterized in that the step
BP neural network in rapid S3 contains two layers of hidden layer, and after test set detection BP neural network reaches required precision, output is every
The weights and threshold value of layer hidden layer, establish dissolved oxygen BP neural network model.
5. a kind of aquaculture water quality dissolved oxygen detection method according to claim 3, which is characterized in that the step S3
In BP neural network contain two layers of hidden layer, when test set detection BP neural network reach required precision after, output every layer it is hidden
The weights and threshold value for hiding layer, establish dissolved oxygen BP neural network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810440821.5A CN108647783A (en) | 2018-05-10 | 2018-05-10 | A kind of aquaculture water quality dissolved oxygen detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810440821.5A CN108647783A (en) | 2018-05-10 | 2018-05-10 | A kind of aquaculture water quality dissolved oxygen detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108647783A true CN108647783A (en) | 2018-10-12 |
Family
ID=63753891
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810440821.5A Withdrawn CN108647783A (en) | 2018-05-10 | 2018-05-10 | A kind of aquaculture water quality dissolved oxygen detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108647783A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583663A (en) * | 2018-12-12 | 2019-04-05 | 中国水产科学研究院渔业机械仪器研究所 | A kind of night water quality dissolved oxygen amount prediction technique suitable for cultivating pool |
| CN109781951A (en) * | 2018-11-29 | 2019-05-21 | 长春市宇驰检测技术有限公司 | A kind of fishpond water quality monitoring system and monitoring method |
| CN109916885A (en) * | 2019-03-26 | 2019-06-21 | 思源电气股份有限公司 | Insulating oil dissolved oxygen content real time on-line detection device |
| CN110672784A (en) * | 2019-10-29 | 2020-01-10 | 饶宾期 | Water body dissolved oxygen detection device based on machine vision |
| CN110889550A (en) * | 2019-11-22 | 2020-03-17 | 江南大学 | Method for predicting dissolved oxygen in different water areas based on transfer learning |
| CN111047073A (en) * | 2019-11-14 | 2020-04-21 | 佛山科学技术学院 | Neural network-based aquaculture water quality prediction method and system |
| CN111487386A (en) * | 2020-03-30 | 2020-08-04 | 江苏大学 | Automatic detection method for water quality parameters of large-area river crab culture pond |
| CN112257932A (en) * | 2020-10-23 | 2021-01-22 | 中国水利水电科学研究院 | Thermal stratification reservoir dissolved oxygen prediction method |
| CN112345473A (en) * | 2020-10-23 | 2021-02-09 | 中国水利水电科学研究院 | Method for identifying dissolved oxygen control factors of thermal stratification reservoir |
-
2018
- 2018-05-10 CN CN201810440821.5A patent/CN108647783A/en not_active Withdrawn
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109781951B (en) * | 2018-11-29 | 2022-01-25 | 长春市宇驰检测技术有限公司 | Fishpond water quality monitoring system and monitoring method |
| CN109781951A (en) * | 2018-11-29 | 2019-05-21 | 长春市宇驰检测技术有限公司 | A kind of fishpond water quality monitoring system and monitoring method |
| CN109583663A (en) * | 2018-12-12 | 2019-04-05 | 中国水产科学研究院渔业机械仪器研究所 | A kind of night water quality dissolved oxygen amount prediction technique suitable for cultivating pool |
| CN109583663B (en) * | 2018-12-12 | 2022-10-14 | 中国水产科学研究院渔业机械仪器研究所 | Night water dissolved oxygen amount prediction method suitable for aquaculture pond |
| CN109916885A (en) * | 2019-03-26 | 2019-06-21 | 思源电气股份有限公司 | Insulating oil dissolved oxygen content real time on-line detection device |
| CN109916885B (en) * | 2019-03-26 | 2024-04-26 | 上海思源光电有限公司 | Real-time online detection device for content of dissolved oxygen in insulating oil |
| CN110672784A (en) * | 2019-10-29 | 2020-01-10 | 饶宾期 | Water body dissolved oxygen detection device based on machine vision |
| CN110672784B (en) * | 2019-10-29 | 2024-06-11 | 华星德安(河北)环保科技有限公司 | Water body dissolved oxygen detection device based on machine vision |
| CN111047073A (en) * | 2019-11-14 | 2020-04-21 | 佛山科学技术学院 | Neural network-based aquaculture water quality prediction method and system |
| CN111047073B (en) * | 2019-11-14 | 2023-04-25 | 佛山科学技术学院 | Aquaculture water quality prediction method and system based on neural network |
| CN110889550A (en) * | 2019-11-22 | 2020-03-17 | 江南大学 | Method for predicting dissolved oxygen in different water areas based on transfer learning |
| CN111487386A (en) * | 2020-03-30 | 2020-08-04 | 江苏大学 | Automatic detection method for water quality parameters of large-area river crab culture pond |
| CN112345473B (en) * | 2020-10-23 | 2021-08-24 | 中国水利水电科学研究院 | Method for identifying dissolved oxygen control factors of thermal stratification reservoir |
| CN112257932B (en) * | 2020-10-23 | 2021-08-03 | 中国水利水电科学研究院 | Thermal stratification reservoir dissolved oxygen prediction method |
| CN112345473A (en) * | 2020-10-23 | 2021-02-09 | 中国水利水电科学研究院 | Method for identifying dissolved oxygen control factors of thermal stratification reservoir |
| CN112257932A (en) * | 2020-10-23 | 2021-01-22 | 中国水利水电科学研究院 | Thermal stratification reservoir dissolved oxygen prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108647783A (en) | A kind of aquaculture water quality dissolved oxygen detection method | |
| CN110476839B (en) | Optimal regulation and control method and system based on fish growth | |
| CN108717586A (en) | A kind of culture environment of aquatic products dissolved oxygen prediction method based on long memory network in short-term | |
| CN107480775A (en) | A kind of dissolved oxygen in fish pond Forecasting Methodology based on data reparation | |
| CN109583663B (en) | Night water dissolved oxygen amount prediction method suitable for aquaculture pond | |
| Consalvey et al. | Monitoring migration and measuring biomass in benthic biofilms: the effects of dark/far-red adaptation and vertical migration on fluorescence measurements | |
| CN107830891B (en) | Data processing method based on aquaculture water pH value multi-parameter data acquisition device | |
| CN113822233B (en) | Method and system for tracking abnormal fishes cultured in deep sea | |
| CN109657200B (en) | Method for determining burst probability of cyanobacterial bloom in lake reservoir | |
| CN109828089A (en) | DBN-BP-based water quality parameter nitrous acid nitrogen online prediction method | |
| CN110402880A (en) | Aquatic environment in-situ immobilization information processing system and method in a kind of aquaculture | |
| CN110910067A (en) | Intelligent regulation and control method and system for live fish transportation water quality by combining deep learning and Q-learning | |
| CN102876816A (en) | Fermentation process statue monitoring and controlling method based on multi-sensor information fusion | |
| CN102124964A (en) | Intelligent management system for mariculture | |
| CN111487386A (en) | Automatic detection method for water quality parameters of large-area river crab culture pond | |
| US9540604B2 (en) | Apparatus and method for monitoring autotroph cultivation | |
| CN101620195A (en) | Method for detecting internal quality of jirou sweet persimmon by smell sensor | |
| CN116091944A (en) | Aquatic ecology investigation method for aquatic vegetation coverage | |
| CN115200647A (en) | Aquaculture environment monitoring management method and system | |
| CN116611000B (en) | Intelligent hairy crab culture environment monitoring system based on machine learning | |
| Lauguico et al. | Indirect measurement of dissolved oxygen based on algae growth factors using machine learning models | |
| Amaliah et al. | Water quality level for shrimp pond at probolinggo area based on fuzzy classification system | |
| CN104458343B (en) | Method for taking water layer by layer in high algae-laden water source places | |
| CN109633113B (en) | Water quality monitoring and early warning method and system based on medaka step-by-step behavior model | |
| CN116957144A (en) | Water quality dissolved oxygen prediction method based on deep learning under complex cultivation environment |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181012 |
|
| WW01 | Invention patent application withdrawn after publication |