CN113642230A - Machine learning-based intelligent control method for adjustable weir of multi-target complex drainage system - Google Patents
Machine learning-based intelligent control method for adjustable weir of multi-target complex drainage system Download PDFInfo
- Publication number
- CN113642230A CN113642230A CN202110780334.5A CN202110780334A CN113642230A CN 113642230 A CN113642230 A CN 113642230A CN 202110780334 A CN202110780334 A CN 202110780334A CN 113642230 A CN113642230 A CN 113642230A
- Authority
- CN
- China
- Prior art keywords
- discrimination
- drainage system
- control
- data
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000010801 machine learning Methods 0.000 title claims abstract description 13
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 22
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims abstract description 10
- 238000012850 discrimination method Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 31
- 238000009826 distribution Methods 0.000 claims description 19
- 239000013598 vector Substances 0.000 claims description 11
- 238000009825 accumulation Methods 0.000 claims description 9
- 230000002265 prevention Effects 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 239000010865 sewage Substances 0.000 claims description 3
- 238000003672 processing method Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 abstract description 8
- 230000000694 effects Effects 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000003860 storage Methods 0.000 description 6
- 238000012544 monitoring process Methods 0.000 description 5
- 230000008094 contradictory effect Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Abstract
The invention provides an intelligent weir adjusting control method for a multi-target complex drainage system based on machine learning. The invention relates to a discrimination state element, a Mahalanobis distance between a total body and a sample, a prior probability and a posterior probability function structure of a control object, and a feedback solver taking a misjudgment rate as an output. Based on machine learning, discrimination, analysis and calculation, the invention selects sensitive data of key state elements such as pipeline initial water level, key point water level flow, pump station forebay water level, most unfavorable point water level and the like, and carries out auxiliary decision on key steps of drainage system control by four machine learning discrimination methods of Bayes method, linear discrimination, linear diagonal discrimination and secondary discrimination. The external data required by the invention is simple and easy to acquire in real time, the training sample can be continuously expanded and accumulated, and the closer the training sample is to the total, the better the judgment effect is. The invention provides an effective method for improving the intelligent control level of the adjustable weir of the complex drainage system, and is beneficial to improving the overall operation level of the complex drainage system.
Description
Technical Field
The invention belongs to the field of municipal engineering, relates to a control method of a drainage system, and particularly relates to an intelligent control method of an adjustable weir based on a multi-target complex drainage system.
Background
In recent years, urban inland inundation prevention and treatment and black and odorous river treatment are becoming the focus of social attention, the construction of corresponding drainage system upgrading and river runoff pollution control engineering is increasing the speed, and related engineering design for upgrading and controlling the pollution of old drainage systems after the original drainage systems are changed is realized. The operation control of the drainage system after the transformation is more complex, and generally relates to a plurality of targets, namely a flood prevention target which mainly aims at improving the safety of the drainage system and realizing the bid improvement and waterlogging prevention and control; secondly, the pollution control aims at reducing initial rainwater pollution and CSO emission. The integral framework of the complex system is that the newly built interception system and the original drainage system form a new drainage system, and the new interception system and the original drainage system need to work together to achieve the engineering target.
In drainage system engineering, a complex relationship of unification and contradiction exists between flood prevention and pollution control. The 'uniformity' is that the two engineering targets need to be realized in the same engineering in a unified way, and must be comprehensively considered; the "contradictory" is that the preconditions and operating modes for the two engineering goals to be achieved do not match completely, and in some cases, there is a certain conflict between the two engineering goals. For example, under certain conditions, in order to achieve the goal of a pollution control project that rainfall of a certain millimeter number does not release the river (for example, initial rainwater of 10mm is controlled), the municipal pump is started after the rainfall of the first 10mm enters the intercepting system; in order to achieve the waterlogging prevention and control goal of improving the drainage capacity of the system, the regulation and storage system is required to be mainly used as a peak clipping regulation and storage volume, namely, the regulation and storage tank is not put into the regulation and storage tank at the initial moment, and the peak clipping is started when the rain peak is waited for.
The general interception point selects a more flexible control form of an adjustable weir. The adjustment of the weir is very important for realizing the complex drainage and the same functions. Through model calculation, if the weir descending time is too early, the storage regulating system can be filled and closed too early, and cannot continuously participate in peak clipping to cause water accumulation in the system. In addition, the water in the shallow pipe network is discharged into the storage system before the runoff peak value, so that the municipal pump cannot be completely started, the drainage capacity of the shallow pipe network cannot be fully utilized, and if the weir descending time is too late, the inflow weir cannot timely participate in the peak clipping effect. Because municipal pump drainage capacity is not enough to deal with the rain peak, if shallow pipe network is full this moment, and the inflow weir does not descend in time, crosses the weir flow not enough, can lead to system ponding. Therefore, it is critical to perform the weir-reducing operation (increase the flow capacity) at an appropriate time to increase the peak-clipping capacity. The difficulty in adjusting the inflow weir, namely, the weir lowering is the judgment of the weir lowering time, so that the invention aims at how to utilize the meter monitoring data of the drainage system to judge the time point of weir lowering operation in real time.
Machine learning, a type of artificial intelligence, can predict future situations based on large amounts of data collected at present. Briefly, machine learning is the use of algorithms to analyze data, learn from it, and make inferences or predictions. In view of the fact that a large number of effective databases which are continuously updated are generated in the actual operation of the drainage system, the data are analyzed through machine learning, and therefore the judgment feasibility of the weir descending time is high. When the prediction model is established, the prediction result of machine learning is compared with the actual result of training data through the establishment of the learning process, and higher judgment accuracy is realized through continuous adjustment.
Disclosure of Invention
The invention overcomes the problems in the prior art, provides an intelligent control method for an adjustable weir of a multi-target complex drainage system based on machine learning, and achieves the purpose of improving the intelligent control level and the integral operation level of the adjustable weir of the complex drainage system.
In order to achieve the above object, the present invention provides a technical solution comprising:
the intelligent control method of the adjustable weir of the multi-target complex drainage system based on machine learning comprises the following steps:
1) obtaining the discrimination state element data of the control object:
selecting state element data: collecting state element data of the drainage system, wherein the state element data comprises a water level value and a flow value of a pipeline at the position of an adjustable weir, a water level of a front pool of a pump station in the drainage system and water accumulation conditions of the most unfavorable points of an area;
carrying out standardization processing on the state element data to eliminate the limitation of dimension and magnitude for subsequent discrimination and solution;
the normalization method is as follows:
assume that the observation matrix for the p-dimensional vector is as follows:
wherein xijIs normalized by the formula
Wherein SijIs a variable xijThe variance of (a) is determined,
the average value of the variable observed values is taken;
2) control idea for distinguishing state element of controlled object
The control targets are as follows: under the premise of ensuring flood control safety, if water is not accumulated in X years, the rainwater at the initial stage of sewage control and interception enters an interception system; the control elements are as follows: adjustable weirs, municipal pumps; the real-time data is solved through discrimination calculation, and if the type belongs to true, the adjustable weir is lowered to the bottom, so that flood prevention safety is ensured; if water is accumulated in an unfavorable place, triggering the municipal pump to start the pump;
(3) solution of discriminant algorithm
a. Distance discrimination solution
Distance discrimination (Mahalanobis distance, Mahalanobis): the mahalanobis distance (mean vector, covariance matrix) from the sample to the population is the closest, and the mahalanobis distance to other populations is certainly farther, and the calculation principle is as follows:
let G be a p-dimensional population, whose distribution mean vector and covariance matrix are:
let x be (x)1，x2，...，xp)’，y＝(y1，y2，...，yp) ' for two samples taken from the population G, assuming Σ > 0(Σ being a positive definite matrix), the squared mahalanobis distance between x, y is defined as: d2(x，y)＝(x-y)′∑-1(x-y) defines the squared mahalanobis distance of x to the population G as: d2(x，G)＝(x-μ)′∑-1(x-μ)
Let two p-dimensional populations G1And G2The mean vectors of the distributions are respectively mu1And mu2The covariance matrix is ∑ respectively1＞0，Σ2Is greater than 0. Extracting n from the two populations1，n2Sample of (2), denoted as x11，x12，…，x1n1And x21，x22，…，x2n2. The existing data set for one-position judgment is marked as x, and the attribution of the x is judged in an attempt mode, and the following judgment rules are provided
When sigma1＝Σ2When sigma is known, the distance d is determined2(x，G2) And d2(x，G1) Are subtracted to obtain
d2(x，G2)-d2(x，G1)＝(x-μ2)′∑-1(x-μ2)-(x-μ1)′∑-1(x-μ1)
The discriminant rule is denoted by W (x):
wherein W (x) is two linear discriminant functions for distance discrimination, and a is a discriminant coefficient.
When sigma1＝Σ2When ∑ unknown
Order to
I.e. derived from the sample
Estimate of Σ, then an estimate of a and W (x) can be derived
Change W (x) in the discriminant rule expressed by W (x) to
The discrimination rule at this time can be obtained.
When sigma1≠Σ2When known, the
Let J (x) be d2(x，G1)-d2(x，G2)
J (x) is a quadratic discriminant function, and the discriminant rule is:
when sigma1≠Σ2When it is unknown
From sample pair
Σ1，Σ2Make an estimation
The estimation and quadratic discriminant function of squared mahalanobis distance can be obtained
The judgment rule is as follows:
distance judgment if multiple populations
Let k p-dimensional population G1，G2，…，GkThe mean vectors of the distributions are respectively mu1，μ2，…，μkThe covariance matrix is ∑ respectively1>0，Σ2>0，…，Σk>0. Extracting n volumes from each of the k populations1，n2，…，nkIs marked as
The existing group of unknown data is marked as x, the attribution of the x is judged, and the judgment rule is
Similar to the distance judgment of two populations, the judgment is also divided into cases
When sigma1＝Σ2＝…＝ΣkWhen ∑ is known
d2(x，Gi)＝(x-μi)′∑-1(x-μi)
Order to
Then there is d2(x，Gi)＝x′∑-1x-2(I′i+ci)，i＝1，2，...，k
Since there is a common quadratic term in each distance, only the linear part of it needs to be considered.
Let WI(x)＝I′ix+ci，i＝1，2，...，k
The decision rule is rewritten as
x∈G1If, if
Scale Wi(x) Is I linear discriminant functions, IiIs a coefficient of discrimination, wherein ciIs a constant term
When sigma1＝Σ2＝…＝ΣkWhen ∑ unknown
Order to
Wherein, i is 1, 2
Deriving mu from the sampleiSigma, so that I can be obtainedi、ciAnd Wi(x) Is estimated by
Wherein, i is 1, 2
When sigma1、Σ2，…，ΣkNot all equal and unknown
Order to
The rule is determined to be
x∈G1If, if
b. Linear discriminant solution
Linear discrimination, assuming that the prior distribution of each data is p-element normal distribution with the same covariance matrix, and obtaining the joint estimation of the covariance matrix from the sample
The judgment to which the data set belongs can be obtained.
c. Linear diagonal solving
At this time, the diagonal matrix serves as an estimate of the covariance matrix.
d. Quadratic discriminant solution
Assuming that the prior distributions of the data groups are p-element normal distributions, but the covariance matrices are not completely the same, the estimates of the covariance matrices are obtained respectively.
e. Bayes discrimination solving
Bayes discriminant (bayes): bayesian discrimination uses a prior probability to describe the prior knowledge, then corrects the prior probability through a sample to obtain a posterior probability, and finally carries out judgment based on the posterior probability. The calculation principle is as follows:
k p-dimensional overall G1, G2, … and Gk are provided, and the probability density function is f1(x)，f2(x)，…，fk(x) In that respect Assuming that the prior probability of sample x from the population Gi is pi (i ═ 1, 2, 3 …, k), then there is p1+p2+…+pk1. According to bayes theory, the posterior probability of sample x from the population Gi is:
when considering the misjudgment cost, a set of all samples that can be judged to be Gi (i is 1, 2, …, k) according to a certain judgment rule is denoted by Ri, and a cost for misjudging the sample x from Gi to Gi is denoted by c (j | i) (i, j is 1, 2, …, k), c (i | i) is 0. The conditional probability of misinterpreting a sample x from Gi as Gi is:
P(j|i)＝P(x∈Ri|x∈Gi)＝∫fi(x) dx is the average misjudgment cost of any judgment rule:
the judgment rule for minimizing the average misjudgment cost ECM is as follows:
x is equal to Gi, if
And if the average misjudgment cost of the sample judgment rule Gi is smaller than the average misjudgment cost of other populations, judging the sample to Gi.
(4) Solution of false rate
Calculating the misjudgment rate: p (j | i) (i ═ 1, 2) represents the probability that a sample originally belonging to the i-th group is erroneously determined as the j-th group, and P (i | j) (j ═ 1, 2) represents the probability that a sample originally belonging to the j-th group is erroneously determined as the i-th group; the misjudgment probability is Err ═ 0.5P (j | i) +0.5P (i | j).
Through the misjudgment rate and the continuous expansion and accumulation of the training samples in the actual operation, a proper judgment method is selected to intelligently control the adjustable weir, and the judgment effect is better when the training samples are closer to the total.
The invention has the beneficial effects that:
after the drainage system is transformed, the realization of drainage waterlogging prevention and pollution control functions are contradictory, and particularly, how to control the adjustable weir during rainstorm is a difficult point of the operation of the complex drainage system. The invention can realize real-time judgment on whether the adjustable weir is opened or not by utilizing the easily-obtained real-time monitoring data and based on a machine learning method, and can ensure the safety of regional drainage and assist manual decision on the premise of realizing pollution control as much as possible. Compared with a model simulation and feedback method, the method has no delay in output time, considers the complexity of external input such as rainfall, rainfall intensity, initial state of a drainage system, production convergence conditions and the like, and compared with the uncertainty of the model simulation method, the accuracy of feedback output by using the 'ash box' model is greatly improved along with the accumulation of the database, so that the method is a 'growing' method. In a word, the external data required by the invention is simple and easy to acquire in real time, the training samples can be continuously expanded and accumulated, and the closer the training samples are to the whole, the better the judgment effect is. The invention provides an effective method for improving the intelligent control level of the adjustable weir of the complex drainage system, and is beneficial to improving the overall operation level of the complex drainage system.
Detailed Description
The following description of the embodiments of the present invention is provided for illustrative purposes, and other advantages and effects of the present invention will become apparent to those skilled in the art from the present disclosure.
Examples
In this embodiment, a certain drainage system is taken as an example, a level meter, a flowmeter, and a camera are used to collect a water level value and a flow value of a pipeline at an adjustable weir position of the drainage system, a water level of a pool in front of a pump station in the drainage system, and a water accumulation condition at the most unfavorable point in an area, and the following variables are selected as a sample matrix (the last data is assumed unknown), wherein an operation category of 1 indicates that a weir descending operation should be performed under the same row of monitoring data, and an operation category of 0 indicates that a weir descending operation should not be performed under the same row of monitoring data.
Table 1: inflow point flow and weir-descending operation truth table of drainage system
Because the dimension and the magnitude of each variable are inconsistent, some are water levels, some are flow rates, the data needs to be standardized firstly, so that the limitation of the dimension and the magnitude is eliminated, and the subsequent statistical analysis is facilitated. The normalization processing method is as follows:
the observation matrix is shown in table 1, which can be written in the form of a matrix (19 × 7):
wherein xijIs normalized by the formula
Wherein SijIs a variable xijThe variance of (a) is determined,
the average value of the variable observed values is taken;
after the normalized formula is changed, table 2 can be obtained.
Table 2: result table after standardized processing of monitoring data samples
The control idea of the distinguishing state element of the control object in the embodiment is as follows: the control target is to realize the purpose that rainwater enters the intercepting system at the initial stage of sewage control and interception on the premise of ensuring flood control safety (no water accumulation in X years). The control elements are as follows: adjustable weir, municipal pump. The real-time data is solved through discrimination calculation, and if the type belongs to true, the adjustable weir is lowered to the bottom, so that flood prevention safety is ensured; if the accumulated water is not favorable, the municipal pump is triggered to be started.
Further, the solution is obtained by using the discrimination algorithm described in the specification of the present invention.
Let G be a p-dimensional population, whose distribution mean vector and covariance matrix are:
can be calculated according to Table 2
Let x be (x)1，x2，...，xp)’，y＝(y1，y2，...，yp) For the data in table 2, assuming Σ > 0(Σ being a positive definite matrix), we define the squared mahalanobis distance between x, y as:
d2(x，y)＝(x-y)′∑-1(x-y) defines the squared mahalanobis distance of x to the population G as: d2(x，G)＝(x-μ)′∑-1(x-μ)；
Let two p-dimensional populations G1And G2The mean vectors of the distributions are respectively mu1And mu2The covariance matrix is ∑ respectively1＞0，Σ2Is greater than 0; extracting n from the two populations1，n2Sample of (2), denoted as x11，x12，…，x1n1And x21，x22，…，x2n2. The existing data set for position judgment is marked as x, and the attribution of x is judged in an attempt mode, wherein the existing data set for position judgment has the following judgment rules:
therefore, whether x needs to be reduced or not can be judged according to whether x belongs to G, wherein the estimation of covariance matrixes of different discrimination modes is as follows:
linear discrimination solving: assuming that the prior distribution of each data is p-element normal distribution with the same covariance matrix, obtaining a joint estimation sigma of the covariance matrix from the sample at the moment, and obtaining the judgment of the data group;
solving linear diagonal: at this time, the diagonal matrix is used as an estimate of the covariance matrix;
and (3) secondary discrimination and solving: assuming that prior distributions of the data groups are p-element normal distributions, but covariance matrixes are not completely the same, and respectively obtaining the estimation of each covariance matrix at the moment;
solving by a Bayes discrimination method: describing the existing knowledge by using a prior probability, then correcting the prior probability by using a sample to obtain a posterior probability, and finally judging based on the posterior probability;
TABLE 3 discrimination result table by different methods
Calculating the misjudgment rate: p (j | i) (i ═ 1, 2) represents the probability that a sample originally belonging to the i-th group is erroneously determined as the j-th group, and P (i | j) (j ═ 1, 2) represents the probability that a sample originally belonging to the j-th group is erroneously determined as the i-th group; the misjudgment probability is Err ═ 0.5P (j | i) +0.5P (i | j). Finally, the misjudgment rate of different algorithm controls of a certain drainage system is obtained. The calculation results are shown in Table 4:
TABLE 4 discrimination Table for different methods
According to the method, the most suitable discrimination method is selected through the misjudgment rate and the continuous expansion and accumulation of the training samples in the actual operation, the adjustable weir is intelligently controlled, and the judgment effect is better when the training samples are closer to the total.
The foregoing merely represents embodiments of the present invention, which are described in some detail and detail, and therefore should not be construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (1)
1. The intelligent control method of the adjustable weir of the multi-target complex drainage system based on machine learning comprises the following steps:
1) obtaining the discrimination state element data of the control object:
selecting state element data: collecting state element data of the drainage system, wherein the state element data comprises a water level value and a flow value of a pipeline at the position of an adjustable weir, a water level of a front pool of a pump station in the drainage system and water accumulation conditions of the most unfavorable points of an area;
carrying out standardization processing on the state element data to eliminate the limitation of dimension and magnitude for subsequent discrimination and solution; the normalization processing method comprises the following steps:
assume that the observation matrix for the p-dimensional vector is as follows:
wherein xijIs normalized by the formula
Wherein SijIs a variable xijVariance of (2)，
The average value of the variable observed values is taken;
2) determining discrimination status element logic of controlled object
The control targets are as follows: under the premise of ensuring flood control safety, if water is not accumulated in X years, the rainwater at the initial stage of sewage control and interception enters an interception system; the control elements are as follows: adjustable weirs, municipal pumps; the real-time data is solved through discrimination calculation, and if the type belongs to true, the adjustable weir is lowered to the bottom, so that flood prevention safety is ensured; if water is accumulated in an unfavorable place, triggering the municipal pump to start the pump;
3) solving a discrimination algorithm, wherein the discrimination algorithm comprises the following steps:
a. distance discrimination solution
The distance discrimination method is that the Mahalanobis distance from the sample to the belonging population, namely the mean vector and the covariance matrix, is nearest; the mahalanobis distance to other populations is somewhat longer, and the calculation principle is as follows:
let G be a p-dimensional population, whose distribution mean vector and covariance matrix are:
let x be (x)1，x2，...，xp)’，y＝(y1，y2，...，yp) ' for two samples taken from the population G, assuming Σ > 0(Σ being a positive definite matrix), the squared mahalanobis distance between x, y is defined as: d2(x，y)＝(x-y)′∑-1(x-y) defines the squared mahalanobis distance of x to the population G as: d2(x，G)＝(x-μ)′∑-1(x-μ)；
Let two p-dimensional populations G1And G2The mean vectors of the distributions are respectively mu1And mu2The covariance matrix is ∑ respectively1＞0，Σ2Is greater than 0; extracting n from the two populations1，n2Sample of (2), denoted as x11，x12，…，x1n1And x21，x22，…，x2n2. The existing data set for position judgment is marked as x, and the attribution of x is judged in an attempt mode, wherein the existing data set for position judgment has the following judgment rules:
b. linear discrimination solving: assuming that the prior distribution of each data is p-element normal distribution with the same covariance matrix, the joint estimation of the covariance matrix is obtained from the samples
The judgment of the data group can be obtained;
c. solving linear diagonal: at this time, the diagonal matrix is used as an estimate of the covariance matrix;
d. and (3) secondary discrimination and solving: assuming that prior distributions of the data groups are p-element normal distributions, but covariance matrixes are not completely the same, and respectively obtaining the estimation of each covariance matrix at the moment;
e. solving by a Bayes discrimination method: describing the existing knowledge by using a prior probability, then correcting the prior probability by using a sample to obtain a posterior probability, and finally judging based on the posterior probability;
4) carrying out misjudgment rate solving calculation to obtain misjudgment rates of different algorithm operations of a certain drainage system; the misjudgment rate calculation is to use P (j | i) (i ═ 1, 2) to represent the probability that the sample originally belonging to the ith group is misjudged as the jth group, and use P (i | j) (j ═ 1, 2) to represent the probability that the sample originally belonging to the jth group is misjudged as the ith group; the misjudgment probability is Err ═ 0.5P (j | i) +0.5P (i | j);
5) and selecting a discrimination algorithm to control the adjustable weir through the misjudgment rate and the expansion and accumulation of the training samples in the actual operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110780334.5A CN113642230B (en) | 2021-07-09 | Multi-target complex drainage system adjustable weir intelligent control method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110780334.5A CN113642230B (en) | 2021-07-09 | Multi-target complex drainage system adjustable weir intelligent control method based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113642230A true CN113642230A (en) | 2021-11-12 |
| CN113642230B CN113642230B (en) | 2024-05-10 |
Family
ID=
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114968647A (en) * | 2022-08-02 | 2022-08-30 | 中海油能源发展股份有限公司采油服务分公司 | Monitoring data deep mining method based on fault diagnosis of flash steam compressor unit |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101806861A (en) * | 2010-04-19 | 2010-08-18 | 哈尔滨工业大学 | Method for judging category of failures caused by electrical contact in sealed electromagnetic relay |
| US7961956B1 (en) * | 2009-09-03 | 2011-06-14 | Thomas Cecil Minter | Adaptive fisher's linear discriminant |
| CN108920428A (en) * | 2018-06-07 | 2018-11-30 | 辽宁工程技术大学 | A kind of fuzzy distance method of discrimination based on joint Fuzzy Extension Principle |
| CN110852364A (en) * | 2019-10-31 | 2020-02-28 | 中国煤炭地质总局勘查研究总院 | Method and device for identifying water source of water burst in mine and electronic equipment |
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7961956B1 (en) * | 2009-09-03 | 2011-06-14 | Thomas Cecil Minter | Adaptive fisher's linear discriminant |
| CN101806861A (en) * | 2010-04-19 | 2010-08-18 | 哈尔滨工业大学 | Method for judging category of failures caused by electrical contact in sealed electromagnetic relay |
| CN108920428A (en) * | 2018-06-07 | 2018-11-30 | 辽宁工程技术大学 | A kind of fuzzy distance method of discrimination based on joint Fuzzy Extension Principle |
| CN110852364A (en) * | 2019-10-31 | 2020-02-28 | 中国煤炭地质总局勘查研究总院 | Method and device for identifying water source of water burst in mine and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 史秀志;周健;郑纬;胡海燕;王怀勇;: "边坡稳定性预测的Bayes判别分析方法及应用", 四川大学学报(工程科学版), no. 03, 20 May 2010 (2010-05-20) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114968647A (en) * | 2022-08-02 | 2022-08-30 | 中海油能源发展股份有限公司采油服务分公司 | Monitoring data deep mining method based on fault diagnosis of flash steam compressor unit |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111222698B (en) | Internet of things-oriented ponding water level prediction method based on long-time and short-time memory network | |
| CN112989538B (en) | Control method and control device for urban drainage system | |
| Lei et al. | Stochastic optimal operation of reservoirs based on copula functions | |
| CN109299812B (en) | Flood prediction method based on deep learning model and KNN real-time correction | |
| WO2021196997A1 (en) | Method for predicting and controlling water level of series of water conveyance canals on basis of fuzzy neural network | |
| Dong et al. | Enhancing short-term forecasting of daily precipitation using numerical weather prediction bias correcting with XGBoost in different regions of China | |
| Wang et al. | Enhancing robustness of monthly streamflow forecasting model using embedded-feature selection algorithm based on improved gray wolf optimizer | |
| CN117035201B (en) | Multi-target scheduling rule making method and system for plain river network water engineering cluster | |
| CN113705877A (en) | Real-time monthly runoff forecasting method based on deep learning model | |
| CN112766603A (en) | Traffic flow prediction method, system, computer device and storage medium | |
| CN111525587A (en) | Reactive load situation-based power grid reactive voltage control method and system | |
| CN114219131A (en) | Watershed runoff prediction method based on LSTM | |
| CN113050430A (en) | Drainage system control method based on robust reinforcement learning | |
| CN112541615A (en) | Water level prediction method based on convolutional neural network | |
| Li et al. | Wind power prediction based on PSO-Kalman | |
| CN113642230A (en) | Machine learning-based intelligent control method for adjustable weir of multi-target complex drainage system | |
| CN113642230B (en) | Multi-target complex drainage system adjustable weir intelligent control method based on machine learning | |
| Badyalina et al. | Streamflow estimation at ungauged basin using modified group method of data handling | |
| CN115689198A (en) | Optimized dispatching method for waterlogging draining pump station in plain polder area | |
| CN111401630B (en) | Flood control system reservoir attribute real-time dynamic evaluation method based on rough set theory | |
| CN110807475B (en) | Flood classification, identification and forecast method based on certainty coefficient | |
| CN114357865A (en) | Hydropower station runoff and associated source load power year scene simulation and prediction method thereof | |
| CN112712159A (en) | LSTM short-time traffic flow prediction method based on improved PSO algorithm | |
| CN112926786A (en) | Shallow lake target water level reverse prediction method and system based on association rule model and numerical simulation | |
| CN112084461B (en) | KNN real-time correction method based on historical flood learning |
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 |