CN114564883B - Lake chlorophyll a short-term set forecasting method and system integrating mechanism and ML - Google Patents
Lake chlorophyll a short-term set forecasting method and system integrating mechanism and ML Download PDFInfo
- Publication number
- CN114564883B CN114564883B CN202210128616.1A CN202210128616A CN114564883B CN 114564883 B CN114564883 B CN 114564883B CN 202210128616 A CN202210128616 A CN 202210128616A CN 114564883 B CN114564883 B CN 114564883B
- Authority
- CN
- China
- Prior art keywords
- chlorophyll
- forecasting
- lake
- algae
- short
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A20/00—Water conservation; Efficient water supply; Efficient water use
- Y02A20/152—Water filtration
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for forecasting a short-term set of lake chlorophyll a by fusing a mechanism and ML. And then establishing a lake algae dynamics mechanism model, and calculating to obtain daily concentration data of the lake chlorophyll a and the key water environment factors. And constructing a time sequence forecasting model set based on machine learning based on time sequence data of chlorophyll a and key factors thereof in daily scale, so as to realize short-term multi-mode forecasting of chlorophyll a in the future days. And finally, establishing a multiple linear regression set forecasting method by combining short-term forecasting results of a plurality of groups of machine learning models, and realizing high-precision short-term set forecasting of the chlorophyll a of the lake. The method effectively solves the problem of lack of data for monitoring the daily chlorophyll a and the water environment factor, and effectively improves the forecasting precision by establishing a multiple linear regression set forecasting method.
Description
Technical Field
The invention relates to the technical field of water conservancy, in particular to a lake chlorophyll a short-term set forecasting method and system integrating a mechanism model and Machine Learning (ML).
Background
Lakes are important components of global water resources, have functions of regulating river runoff, supplying water and irrigation, improving ecological environment, maintaining balance of an ecological system and the like, and have important influence on human survival and development. In recent years, lake eutrophication has become a serious challenge worldwide with the dual effects of increased human activity and global warming. The outbreak of cyanobacterial bloom caused by lake eutrophication has become a serious ecological environment problem faced for a long time in the world at present and in the future. The dynamic monitoring, forecasting and early warning research of the lake eutrophication is developed, is the basis for preventing and treating the lake algal bloom, and has important significance for controlling the lake eutrophication and improving the water ecological environment.
Chlorophyll a concentration is closely related to abundance and biomass of aquatic phytoplankton, and is a common index for evaluating eutrophication and algal bloom. The main modes for obtaining chlorophyll a concentration at present include: site sampling investigation, remote sensing satellite observation and model simulation. The chlorophyll a data obtained by field sampling has the highest precision, and laboratory analysis is mainly carried out by manual sampling every month, but the method is time-consuming and labor-consuming, and the spatial up-sampling sites are sparse. Satellite remote sensing technology can perform inversion of chlorophyll a in a large range in space, but the accuracy is relatively low, and the data quality is seriously disturbed by the atmosphere. Chlorophyll a high frequency data of long time series can be obtained through model simulation, but the model often cannot provide future forecasting results due to lack of future model input boundary conditions. At present, short-term forecasting research of the lake chlorophyll a mainly carries out monitoring data acquisition in a short time scale (an hour scale or a day scale) by arranging related high-frequency monitoring instruments and equipment, and then forecasting is carried out by combining a related statistical method, a machine learning method and the like. For lakes lacking high frequency observations, there is currently no effective short-term prediction utility. In addition, in the current research, the prediction mode is single, and the uncertainty of the prediction result is large.
Disclosure of Invention
The invention aims to: the invention provides a method and a system for forecasting short-term collection of chlorophyll a in a lake by combining a mechanism model and machine learning, which provide important technical support for accurate dynamic monitoring, forecasting and early warning and scientific management and control of eutrophication of the lake.
The technical scheme is as follows: in order to achieve the aim of the invention, the invention adopts the following technical scheme:
a lake chlorophyll a short-term set forecasting method integrating a mechanism model and machine learning ML comprises the following steps:
carrying out correlation analysis on the lake chlorophyll a concentration data with a month scale and meteorological and water environment influence factors thereof, and identifying key factors influencing chlorophyll a;
establishing a lake algae dynamics mechanism model, performing calibration verification by using month scale data, and calculating to obtain daily concentration data of lake chlorophyll a and key water environment factors;
constructing a time sequence forecasting model set based on machine learning based on time sequence data of chlorophyll a and key meteorological factors and water environment factors thereof in a daily scale, wherein the time sequence forecasting model set comprises SVM, RF and LSTM models, and realizing short-term multi-mode forecasting of chlorophyll a in the future days;
and establishing a multiple linear regression set forecasting method by combining short-term forecasting results of the lake chlorophyll a of a plurality of groups of machine learning models, so as to realize short-term set forecasting of the lake chlorophyll a.
Preferably, the water quality variable equation in the lake algae dynamics mechanism model is:
wherein C is i Is the concentration of water quality variable; u, v, w are velocity components in the x, y, z directions under the horizontal-curve coordinate and the vertical coordinate respectively; h is the water depth; a is that x ,A y ,A z Turbulence diffusion coefficients in the x, y and z directions respectively;an internal and external source sink for each unit of water volume; m is m x 、m y Is a horizontal-curve coordinate variation factor; t is time;
the temperature equation is:
wherein T is the temperature; a is that b Is a vertical turbulent diffusion coefficient; i is the intensity of solar shortwave radiation; s is S T Is the source sink of the heat exchange.
The dynamic process equation of algae in the lake algae dynamic mechanism model is as follows:
wherein B is biomass of algae, P is growth rate of algae, BM is basal metabolism rate of algae, PR is predation rate of algae, WS is sedimentation rate of algae, WB is exogenous load of algae, and V is unit volume.
Preferably, before constructing a time sequence forecasting model set based on machine learning, carrying out normalization pretreatment on the time sequence data of a day scale, wherein chlorophyll a time sequence data is taken as an output item y, meteorological and water environment key influence factors are taken as an input item vector x, and the data value ranges of y and x are converted into [0,1]]The method comprises the steps of carrying out a first treatment on the surface of the Dividing the data set of the whole time sequence into a training data set and a test data set, determining the data corresponding relation according to the future days to be predicted, and predicting the result of 1 day, then x t-1 →y t The method comprises the steps of carrying out a first treatment on the surface of the Predicting the result of 2 days, then x t-2 →y t The method comprises the steps of carrying out a first treatment on the surface of the Predicting the result of 3 days, then x t-3 →y t The method comprises the steps of carrying out a first treatment on the surface of the t is time.
Preferably, the set forecast result F of the lake chlorophyll a is calculated according to the following formula:
wherein N is the number of machine learning models, M i Forecasting result for ith machine learning model, alpha i Beta is the corresponding weight coefficient 0 Is constant term beta 0 。
Based on the same inventive concept, the invention provides a lake chlorophyll a short-term set forecasting system integrating a mechanism model and machine learning, which comprises the following steps:
the key influence factor identification module is used for carrying out correlation analysis on the concentration data of chlorophyll a in the lake with a month scale and meteorological and water environment influence factors and identifying key factors influencing chlorophyll a;
the daily scale data acquisition module is used for establishing a lake algae dynamics mechanism model, performing calibration verification by using monthly scale data, and calculating to obtain daily concentration data of the lake chlorophyll a and the key water environment factors;
the machine learning module is used for constructing a time sequence forecasting model set based on machine learning and comprising SVM, RF and LSTM models based on time sequence data of chlorophyll a and key meteorological and water environment factors thereof in daily scale, so as to realize short-term multi-mode forecasting of chlorophyll a in the future days;
and the set forecasting module is used for combining short-term forecasting results of the lake chlorophyll a of the multiple groups of machine learning models, establishing a multiple linear regression set forecasting method and realizing short-term set forecasting of the lake chlorophyll a.
Based on the same inventive concept, the invention also provides a computer system, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program realizes the fusion mechanism model and the short-term set forecasting method of the machine learning lake chlorophyll a when being loaded to the processor.
The beneficial effects are that: because of the lack of daily-scale chlorophyll a monitoring data, short-term prediction of the lake chlorophyll a is less concerned in the related researches in the past, and part of few researches mainly use high-frequency automatic monitoring station observation data to perform single-mode lake chlorophyll a prediction of statistical and machine learning algorithms, but the prediction effect is poor due to a plurality of uncertainties of a single model. Compared with the prior art, the invention provides an effective short-term aggregate prediction calculation method for the lake chlorophyll a, which can not only effectively overcome the problem of lack of daily scale monitoring data, but also remarkably improve the short-term prediction precision of the lake chlorophyll a by adopting the idea of integrated learning.
Drawings
FIG. 1 is a schematic flow diagram of one embodiment of the present invention;
FIG. 2 is a schematic diagram of a model-based modeling of the modeling of chlorophyll a concentration;
FIG. 3 is a schematic diagram of test data set and training data set partitioning;
FIG. 4 is a schematic diagram of aggregate forecast results.
Detailed Description
The technical scheme of the invention is described in detail below with reference to the accompanying drawings and specific embodiments.
As shown in fig. 1, the method for forecasting the short-term set of chlorophyll a in a lake by fusing a mechanism model and machine learning disclosed by the embodiment of the invention comprises the following steps:
(1) And collecting the concentration data of chlorophyll a in the lake with a month scale, analyzing the pearson correlation between the chlorophyll a and the influence factors, and identifying key factors influencing the chlorophyll a.
The method specifically comprises the following steps:
(1-1) month scale data collection: acquisition of chlorophyll a data, meteorological data (rainfall, illumination, wind speed, wind direction and the like) and water environment data (water temperature, dissolved oxygen, total nitrogen, total phosphorus and the like) of a lake site on a month scale for a plurality of years is performed.
(1-2) key impact factor identification: and respectively calculating pearson correlation coefficients between the lake chlorophyll a data and each factor, and selecting the significance level p <0.05 of the factor correlation analysis result as a key influence factor so as to identify key weather and water environment factors influencing the lake chlorophyll a.
(2) Collecting daily scale data of key meteorological factors; and (3) establishing a lake algae dynamics mechanism model, and calculating to obtain daily concentration data of the lake chlorophyll a and the key water environment factors.
The method specifically comprises the following steps:
daily data of key meteorological factors are obtained from a meteorological monitoring department, basic data such as hydrology, topography and the like of a research area are collected, an open source model EFDC is adopted to build a lake algae dynamics mechanism model, daily chlorophyll a and key water environment factors (water temperature, dissolved oxygen, total nitrogen, total phosphorus and the like) are simulated and output, and calibration is carried out according to actual measurement data of each month (as shown in figure 2). Finally, the daily scale data of the chlorophyll a in the lake and the daily scale data of key influence factors thereof are obtained. The lake algae dynamics mechanism model control equation specifically comprises:
A. water quality variable equation:
wherein C is i The concentration of a certain water quality variable is in mg/L; u, v, w are velocity components in the x, y and z directions under the horizontal-curve coordinate and the vertical coordinate respectively, and the unit is m/s; h is the depth of water, and the unit is m; a is that x ,A y ,A z Turbulence diffusion coefficients in the x, y and z directions respectively;is an internal and external source and sink of water per unit water volume, and the unit is g/m 3 ;m x 、m y Is a horizontal-curve coordinate variation factor; t is time in s.
B. Temperature equation:
wherein T is the temperature in the unit of DEG C; a is that b Is a vertical turbulent diffusion coefficient, and has the unit of m 2 S; i is the intensity of solar shortwave radiation, and the unit is W/m 2 ;S T Is the source sink of heat exchange, and is expressed as J/s.
C. Algae (chlorophyll a) dynamic process equation:
wherein B is biomass (chlorophyll a concentration) of algae in g/m 3 The method comprises the steps of carrying out a first treatment on the surface of the t is time, and the unit is d; p is the growth rate of algae in d -1 The method comprises the steps of carrying out a first treatment on the surface of the BM is the basal metabolic rate of algae in d -1 The method comprises the steps of carrying out a first treatment on the surface of the PR is the predation rate of algae in d -1 The method comprises the steps of carrying out a first treatment on the surface of the WS is the sedimentation rate of algae, and the unit is m/d; WB is the exogenous load of algae in g/d; v is the unit volume, the unit is m 3 。
(3) Based on time series data of chlorophyll of a daily scale and key meteorological factors and water environment factors thereof, a time series forecasting model set (support vector machine SVM, random forest RF and long-term memory artificial neural network LSTM) based on machine learning is constructed, and short-term multi-mode forecasting of chlorophyll a in the future days is realized.
The method specifically comprises the following steps:
and (3-1) carrying out normalization pretreatment on the time series data of the day scale, wherein the time series data of chlorophyll a is an output item y, a weather and water environment key influence factor is an input item vector x (time series data containing i components), and the data value ranges of y and x are converted into [0,1], and the specific forms are as follows:
wherein a is * A is the normalized result of the data max 、a min Respectively, the maximum value and the minimum value of the time series data.
(3-2) the entire time series is of length nThe data set, divided into training data set (1-t) and test data set (t-n), is shown in FIG. 3. Predicting 1 day result according to the data corresponding relation determined by the future days to be predicted, and x t-1 →y t The method comprises the steps of carrying out a first treatment on the surface of the Predicting the result of 2 days, then x t-2 →y t The method comprises the steps of carrying out a first treatment on the surface of the Predicting the result of 3 days, then x t-3 →y t . According to the data correspondence, and so on, a machine learning model set (SVM, RF and LSTM) is constructed to train model parameters of the training data set, and parameter calibration verification is carried out according to the test data set to determine optimal model parameters (as shown in table 1).
Table 1 machine learning model parameter set
(3-3) output results m for 3 machine learning model predictions i (i=1, 2, 3) to obtain a true predictive value of chlorophyll a.
M i =m i ×(a max -a min )+a min
Wherein a is max 、a min Maximum value and minimum value of chlorophyll a data of original day scale time sequence respectively, M i And (5) performing inverse normalization on the i-th machine learning model prediction.
(4) And establishing a multiple linear regression set forecasting method by combining short-term forecasting results of the lake chlorophyll a of a plurality of groups of machine learning models, so as to realize high-precision short-term set forecasting of the lake chlorophyll a.
The method specifically comprises the following steps:
forecasting the result M by using limited number of lake chlorophyll a observations and corresponding 3 machine learning models i Multiple linear regression is carried out, and the weight coefficient alpha corresponding to each mode is solved according to the least square principle i And a constant term beta 0 . And further calculating to obtain a set forecasting result F of the lake chlorophyll a according to the following formula.
Fig. 4 shows the aggregate forecast result of chlorophyll a of the large Gong Shan site, and it can be seen from the figure that the aggregate forecast is better than that of a single machine learning model, the forecast result is closer to a reference value, and the accuracy and reliability of the aggregate forecast result are strong at different forecast times.
Based on the same inventive concept, the embodiment of the invention provides a lake chlorophyll a short-term set forecasting system integrating a mechanism model and machine learning, which comprises the following steps: the key influence factor identification module is used for carrying out correlation analysis on the concentration data of chlorophyll a in the lake with a month scale and meteorological and water environment influence factors and identifying key factors influencing chlorophyll a; the daily scale data acquisition module is used for establishing a lake algae dynamics mechanism model, performing calibration verification by using monthly scale data, and calculating to obtain daily concentration data of the lake chlorophyll a and the key water environment factors; the machine learning module is used for constructing a time sequence forecasting model set based on machine learning ML (model support vector machine), RF (radio frequency) and LSTM (least squares) models based on time sequence data of chlorophyll a and key meteorological and water environment factors thereof in daily scale, so as to realize short-term multi-mode forecasting of chlorophyll a in the future days; and the set forecasting module is used for combining short-term forecasting results of the lake chlorophyll a of the multiple groups of machine learning models, establishing a multiple linear regression set forecasting method and realizing short-term set forecasting of the lake chlorophyll a.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of each module described above may refer to the corresponding process in the foregoing method embodiment, which is not described herein again. The division of the modules is merely a logic function division, and other division manners may be implemented in practice, for example, multiple modules may be combined or may be integrated into another system.
Based on the same inventive concept, the embodiment of the invention also provides a computer system, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program realizes the fusion mechanism model and the short-term set forecasting method of the machine-learned lake chlorophyll a when being loaded to the processor.
It will be appreciated by those skilled in the art that aspects of the present invention, in essence or contributing to the prior art, may be embodied in the form of a software product stored in a storage medium, comprising instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of the embodiments of the present invention. The storage medium includes: a usb disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk, or other various media capable of storing a computer program.
Claims (8)
1. A lake chlorophyll a short-term set forecasting method integrating a mechanism and ML is characterized by comprising the following steps:
carrying out correlation analysis on the lake chlorophyll a concentration data with a month scale and meteorological and water environment influence factors thereof, and identifying key factors influencing chlorophyll a;
establishing a lake algae dynamics mechanism model, performing calibration verification by using month scale data, and calculating to obtain daily concentration data of lake chlorophyll a and key water environment factors;
constructing a time sequence forecasting model set based on machine learning ML based on time sequence data of chlorophyll a and key meteorological and water environment factors thereof in daily scale, wherein the time sequence forecasting model set comprises SVM, RF and LSTM models, and realizing short-term multi-mode forecasting of chlorophyll a in future days, two days or three days;
combining short-term prediction results of the lake chlorophyll a of a plurality of groups of machine learning models, establishing a multiple linear regression set prediction method, and realizing short-term set prediction of the lake chlorophyll a; forecasting result M through limited times of lake chlorophyll a observation values and corresponding machine learning models i Multiple linear regression is carried out, and the weight coefficient alpha corresponding to each mode is solved according to the least square principle i And a constant term beta 0 The method comprises the steps of carrying out a first treatment on the surface of the Calculated according to the following formulaObtaining a set forecasting result F of the chlorophyll a in the lake:
wherein N is the number of machine learning models, M i Forecasting results for the ith machine learning model.
2. The method for forecasting the short-term set of chlorophyll a in lakes by combining a mechanism and ML according to claim 1, wherein the equation of water quality variables in the model of the dynamics mechanism of the lake algae is:
wherein C is i Is the concentration of water quality variable; u, v, w are velocity components in the x, y, z directions under the horizontal-curve coordinate and the vertical coordinate respectively; h is the water depth; a is that x ,A y ,A z Turbulence diffusion coefficients in the x, y and z directions respectively; s is S Ci An internal and external source sink for each unit of water volume; m is m x 、m y Is a horizontal-curve coordinate variation factor; t is time;
the temperature equation is:
wherein T is the temperature; a is that b Is a vertical turbulent diffusion coefficient; i is the intensity of solar shortwave radiation; s is S T Is the source sink of the heat exchange.
3. The method for forecasting the short-term set of chlorophyll a in lakes by combining a mechanism and ML according to claim 1, wherein an equation of an alga dynamic process in a model of a lake alga dynamic mechanism is:
wherein B is biomass of algae, P is growth rate of algae, BM is basal metabolism rate of algae, PR is predation rate of algae, WS is sedimentation rate of algae, WB is exogenous load of algae, t is time, z is vertical coordinate, and V is unit volume.
4. The method for forecasting the short-term set of chlorophyll a in lakes by combining a mechanism and ML according to claim 1, wherein before a time series forecasting model set based on machine learning is constructed, normalization preprocessing is carried out on time series data of a day scale, wherein the time series data of chlorophyll a is an output item y, meteorological and water environment key influence factors are input item vectors x, and data value fields of y and x are converted into [0,1]]The method comprises the steps of carrying out a first treatment on the surface of the Dividing the data set of the whole time sequence into a training data set and a test data set, determining the data corresponding relation according to the future days to be predicted, and predicting the result of 1 day, then x t-1 →y t The method comprises the steps of carrying out a first treatment on the surface of the Predicting the result of 2 days, then x t-2 →y t The method comprises the steps of carrying out a first treatment on the surface of the Predicting the result of 3 days, then x t-3 →y t The method comprises the steps of carrying out a first treatment on the surface of the t is time.
5. A lake chlorophyll a short-term aggregate forecasting system that fuses a mechanism and ML, comprising:
the key influence factor identification module is used for carrying out correlation analysis on the concentration data of chlorophyll a in the lake with a month scale and meteorological and water environment influence factors and identifying key factors influencing chlorophyll a;
the daily scale data acquisition module is used for establishing a lake algae dynamics mechanism model, performing calibration verification by using monthly scale data, and calculating to obtain daily concentration data of the lake chlorophyll a and the key water environment factors;
the machine learning module is used for constructing a time sequence forecasting model set based on machine learning ML (model support vector machine), RF (radio frequency) and LSTM (least squares) models based on time sequence data of chlorophyll a and key meteorological and water environment factors thereof in daily scale, and realizing short-term multi-mode forecasting of chlorophyll a in the future in one, two or three days;
the set forecasting module is used for combining short-term forecasting results of the lake chlorophyll a of the multiple groups of machine learning models, establishing a multiple linear regression set forecasting method and realizing short-term set forecasting of the lake chlorophyll a; forecasting result M through limited times of lake chlorophyll a observation values and corresponding machine learning models i Multiple linear regression is carried out, and the weight coefficient alpha corresponding to each mode is solved according to the least square principle i And a constant term beta 0 The method comprises the steps of carrying out a first treatment on the surface of the And calculating to obtain a set forecasting result F of the lake chlorophyll a according to the following formula:
wherein N is the number of machine learning models, M i Forecasting results for the ith machine learning model.
6. The system for short-term aggregation prediction of chlorophyll a in lakes with ML and fusion mechanisms according to claim 5, wherein the equation of water quality variables in the model of the dynamics mechanism of lake algae is:
wherein C is i Is the concentration of water quality variable; u, v, w are velocity components in the x, y, z directions under the horizontal-curve coordinate and the vertical coordinate respectively; h is the water depth; a is that x ,A y ,A z Turbulence diffusion coefficients in the x, y and z directions respectively;an internal and external source sink for each unit of water volume; m is m x 、m y Is a horizontal-curve coordinate variation factor; t is time;
the temperature equation is:
wherein T is the temperature; a is that b Is a vertical turbulent diffusion coefficient; i is the intensity of solar shortwave radiation; s is S T Is the source sink of the heat exchange.
7. The system for short-term aggregate forecasting of chlorophyll a in lakes with integrated mechanisms and ML according to claim 5, wherein the equation of the dynamic process of algae in the model of the dynamic mechanism of algae in lakes is:
wherein B is biomass of algae, P is growth rate of algae, BM is basal metabolism rate of algae, PR is predation rate of algae, WS is sedimentation rate of algae, WB is exogenous load of algae, t is time, z is vertical coordinate, and V is unit volume.
8. A computer system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the computer program when loaded into the processor implements a method of short term aggregation forecasting of lake chlorophyll a of ML with a fusion mechanism according to any one of claims 1-4.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210128616.1A CN114564883B (en) | 2022-02-11 | 2022-02-11 | Lake chlorophyll a short-term set forecasting method and system integrating mechanism and ML |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210128616.1A CN114564883B (en) | 2022-02-11 | 2022-02-11 | Lake chlorophyll a short-term set forecasting method and system integrating mechanism and ML |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114564883A CN114564883A (en) | 2022-05-31 |
CN114564883B true CN114564883B (en) | 2023-05-19 |
Family
ID=81712954
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210128616.1A Active CN114564883B (en) | 2022-02-11 | 2022-02-11 | Lake chlorophyll a short-term set forecasting method and system integrating mechanism and ML |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114564883B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20220198303A1 (en) * | 2019-04-15 | 2022-06-23 | Riken | Device, method and program for environmental factor estimation, learned model and recording medium |
CN115526038B (en) * | 2022-09-19 | 2023-11-21 | 国家海洋环境预报中心 | Offshore ecological environment numerical forecasting method and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102135531A (en) * | 2010-12-24 | 2011-07-27 | 中国科学院南京地理与湖泊研究所 | Method for forecasting blue-green algae water bloom in large-scale shallow lake within 72 hours |
CN106990216A (en) * | 2017-04-01 | 2017-07-28 | 河海大学 | A kind of shallow lake wawter bloom risk analysis early warning system and its analysis and early warning method |
CN113642268A (en) * | 2021-07-15 | 2021-11-12 | 中国环境科学研究院 | River water bloom prediction method and system |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018214190A1 (en) * | 2017-06-19 | 2018-11-29 | 中国科学院南京地理与湖泊研究所 | Stereoscopic monitoring and data mining system and method for harmful lake cyanobacteria bloom |
US20210348211A1 (en) * | 2018-04-28 | 2021-11-11 | Nanjing Institute Of Geography & Limnology. Chinese Academy Of Sciences | Method for defining stages of development of cyanobacterial bloom |
CN108846517B (en) * | 2018-06-12 | 2021-03-16 | 清华大学 | Integration method for predicating quantile probabilistic short-term power load |
-
2022
- 2022-02-11 CN CN202210128616.1A patent/CN114564883B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102135531A (en) * | 2010-12-24 | 2011-07-27 | 中国科学院南京地理与湖泊研究所 | Method for forecasting blue-green algae water bloom in large-scale shallow lake within 72 hours |
CN106990216A (en) * | 2017-04-01 | 2017-07-28 | 河海大学 | A kind of shallow lake wawter bloom risk analysis early warning system and its analysis and early warning method |
CN113642268A (en) * | 2021-07-15 | 2021-11-12 | 中国环境科学研究院 | River water bloom prediction method and system |
Also Published As
Publication number | Publication date |
---|---|
CN114564883A (en) | 2022-05-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114564883B (en) | Lake chlorophyll a short-term set forecasting method and system integrating mechanism and ML | |
CN102183621B (en) | Aquaculture dissolved oxygen concentration online forecasting method and system | |
CN104091216A (en) | Traffic information predication method based on fruit fly optimization least-squares support vector machine | |
CN107292436A (en) | Blue-green alga bloom Forecasting Methodology based on nonlinear kinetics temporal model | |
Liu et al. | Prediction of dissolved oxygen content in aquaculture of Hyriopsis cumingii using Elman neural network | |
Li et al. | A novel combined prediction model for monthly mean precipitation with error correction strategy | |
Yu et al. | A hybrid intelligent soft computing method for ammonia nitrogen prediction in aquaculture | |
Wang et al. | Reliable model of reservoir water quality prediction based on improved ARIMA method | |
Jiayu et al. | Application of intelligence information fusion technology in agriculture monitoring and early-warning research | |
Yu et al. | A deep learning approach for multi-depth soil water content prediction in summer maize growth period | |
CN109934422A (en) | Neural network wind speed prediction method based on time series data analysis | |
Zhang et al. | Study on water quality prediction of urban reservoir by coupled CEEMDAN decomposition and LSTM neural network model | |
Xiaoyi et al. | Research on hybrid mechanism modeling of algal bloom formation in urban lakes and reservoirs | |
Wang et al. | An approach of recursive timing deep belief network for algal bloom forecasting | |
CN103675006A (en) | Least-squares-based industrial melt index soft measuring meter and method | |
CN116595333B (en) | Soil-climate intelligent rice target yield and nitrogen fertilizer consumption determination method | |
CN116628442B (en) | Groundwater reserve change space downscaling method based on artificial neural network | |
Wu et al. | WOA-LSTM | |
CN111126827A (en) | Input-output accounting model construction method based on BP artificial neural network | |
Xia et al. | Environmental factor assisted chlorophyll-a prediction and water quality eutrophication grade classification: A comparative analysis of multiple hybrid models based on a SVM | |
Niu et al. | Soil Moisture Prediction in Peri-urban Beijing, China: Gene Expression Programming Algorithm. | |
Zhu et al. | Prediction model of dissolved oxygen based on FOA-LSSVR | |
Jiang et al. | Application and evaluation of an improved LSTM model in the soil moisture prediction of southeast chinese tobacco-producing areas | |
Deng et al. | Estimation of a five-parameter JONSWAP spectra with an improved particle swarm optimization | |
Shuaishuai et al. | Prediction model of transpiration rate of strawberry in closed cultivation based on DBN-LSSVM algorithm |
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 |