CN113096087A - Soft measurement method for purity of phosphorus recovery product - Google Patents
Soft measurement method for purity of phosphorus recovery product Download PDFInfo
- Publication number
- CN113096087A CN113096087A CN202110366780.1A CN202110366780A CN113096087A CN 113096087 A CN113096087 A CN 113096087A CN 202110366780 A CN202110366780 A CN 202110366780A CN 113096087 A CN113096087 A CN 113096087A
- Authority
- CN
- China
- Prior art keywords
- hidden node
- matrix
- model
- purity
- phosphorus recovery
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 229910052698 phosphorus Inorganic materials 0.000 title claims abstract description 45
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 title claims abstract description 42
- 239000011574 phosphorus Substances 0.000 title claims abstract description 42
- 238000011084 recovery Methods 0.000 title claims abstract description 39
- 238000000691 measurement method Methods 0.000 title claims abstract description 10
- 238000000034 method Methods 0.000 claims abstract description 40
- 239000002245 particle Substances 0.000 claims abstract description 25
- 238000005259 measurement Methods 0.000 claims abstract description 18
- 238000004458 analytical method Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000036961 partial effect Effects 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims abstract description 4
- 230000003287 optical effect Effects 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 53
- 238000010238 partial least squares regression Methods 0.000 claims description 14
- 230000003190 augmentative effect Effects 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 230000009471 action Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 3
- 230000002708 enhancing effect Effects 0.000 claims description 3
- 238000003064 k means clustering Methods 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims description 3
- 230000002159 abnormal effect Effects 0.000 claims description 2
- 238000003860 storage Methods 0.000 claims description 2
- 239000000047 product Substances 0.000 description 41
- CKMXBZGNNVIXHC-UHFFFAOYSA-L ammonium magnesium phosphate hexahydrate Chemical compound [NH4+].O.O.O.O.O.O.[Mg+2].[O-]P([O-])([O-])=O CKMXBZGNNVIXHC-UHFFFAOYSA-L 0.000 description 11
- 229910052567 struvite Inorganic materials 0.000 description 11
- 230000008569 process Effects 0.000 description 6
- 239000007787 solid Substances 0.000 description 6
- 238000004519 manufacturing process Methods 0.000 description 5
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 4
- 239000002244 precipitate Substances 0.000 description 4
- 229910019142 PO4 Inorganic materials 0.000 description 3
- 238000002425 crystallisation Methods 0.000 description 3
- 230000008025 crystallization Effects 0.000 description 3
- 239000003337 fertilizer Substances 0.000 description 3
- 150000002500 ions Chemical class 0.000 description 3
- 238000006467 substitution reaction Methods 0.000 description 3
- 238000005033 Fourier transform infrared spectroscopy Methods 0.000 description 2
- 238000002441 X-ray diffraction Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 229910052757 nitrogen Inorganic materials 0.000 description 2
- 239000010452 phosphate Substances 0.000 description 2
- NBIIXXVUZAFLBC-UHFFFAOYSA-K phosphate Chemical compound [O-]P([O-])([O-])=O NBIIXXVUZAFLBC-UHFFFAOYSA-K 0.000 description 2
- 239000002351 wastewater Substances 0.000 description 2
- 241000282412 Homo Species 0.000 description 1
- 241000282414 Homo sapiens Species 0.000 description 1
- FYYHWMGAXLPEAU-UHFFFAOYSA-N Magnesium Chemical compound [Mg] FYYHWMGAXLPEAU-UHFFFAOYSA-N 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000001506 calcium phosphate Substances 0.000 description 1
- 229910000389 calcium phosphate Inorganic materials 0.000 description 1
- 235000011010 calcium phosphates Nutrition 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000004090 dissolution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000016507 interphase Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 239000011777 magnesium Substances 0.000 description 1
- 229910052749 magnesium Inorganic materials 0.000 description 1
- GVALZJMUIHGIMD-UHFFFAOYSA-H magnesium phosphate Chemical compound [Mg+2].[Mg+2].[Mg+2].[O-]P([O-])([O-])=O.[O-]P([O-])([O-])=O GVALZJMUIHGIMD-UHFFFAOYSA-H 0.000 description 1
- 239000004137 magnesium phosphate Substances 0.000 description 1
- 229910000157 magnesium phosphate Inorganic materials 0.000 description 1
- 229960002261 magnesium phosphate Drugs 0.000 description 1
- 235000010994 magnesium phosphates Nutrition 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002367 phosphate rock Substances 0.000 description 1
- OJMIONKXNSYLSR-UHFFFAOYSA-N phosphorous acid Chemical compound OP(O)O OJMIONKXNSYLSR-UHFFFAOYSA-N 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000013268 sustained release Methods 0.000 description 1
- 239000012730 sustained-release form Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- QORWJWZARLRLPR-UHFFFAOYSA-H tricalcium bis(phosphate) Chemical compound [Ca+2].[Ca+2].[Ca+2].[O-]P([O-])([O-])=O.[O-]P([O-])([O-])=O QORWJWZARLRLPR-UHFFFAOYSA-H 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/64—Analysis of geometric attributes of convexity or concavity
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/20—Identification of molecular entities, parts thereof or of chemical compositions
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C20/00—Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
- G16C20/70—Machine learning, data mining or chemometrics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Geometry (AREA)
- Chemical & Material Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Crystallography & Structural Chemistry (AREA)
- Evolutionary Biology (AREA)
- Spectroscopy & Molecular Physics (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A soft measurement method for the purity of a phosphorus recovery product comprises the following steps: (1) shooting the collected phosphorus recovery product by using an optical microscope, and recording the micro-morphology structure of the phosphorus recovery product on a two-dimensional plane; (2) processing the picture shot in the step (1); (3) measurement and calculation: distance nF between two points farthest from the periphery of the particlemax(ii) a Distance nF between two nearest points on the periphery of the particlemin(ii) a Elongation nEl1(ii) a Aspect ratio nEl2(ii) a Total area n of particlesA(ii) a Perimeter n of particleP(ii) a Perimeter n of particleCi(ii) a Equivalent diameter nEd(ii) a Solidity ns(ii) a Roughness nCo(ii) a (4) Establishing a soft measurement model of the purity of the phosphorus recovery product by adopting a nonlinear partial least square NLPLS method; and for a new phosphorus recovery product, performing predictive analysis on the product purity by using an NLPLS model, and updating the model by using new predictive analysis data. The invention provides a rapid, simple and accurate method for determining the purity of the phosphorus recovery product.
Description
Technical Field
The invention belongs to the technical field of environmental protection, and particularly relates to a soft measurement method for the purity of a phosphorus recovery product.
Background
Phosphorus is an important nutrient element and has important significance for human beings and various organisms in the natural world. Due to the large amount of mining by humans and the unidirectional flow of phosphorus in nature, phosphorite will be mined out within 50 to 100 years in the future, and at that time a serious phosphorus resource crisis will result. To alleviate this phenomenon, techniques for recovering and reusing phosphorus resources have been receiving wide attention. In the last decade, struvite crystallization has been considered as a very promising approach to solve the phosphorus resource crisis, and not only can precipitate phosphate in wastewater, but also the product MgNH4PO3 · H2O (struvite) is a good slow-release fertilizer, and thus has received extensive attention. The crystallization process is comprehensively influenced by liquid-solid equilibrium thermodynamics and solid-liquid interphase mass transfer, and other process factors, such as pH, supersaturation degree, mixing energy, temperature and the existence of competitive ions, can influence the purity of the product struvite. High strength, low dissolution rate, sustained release properties and high purity are all of the primary considerations in order for struvite to be an economically viable commercial fertilizer.
In general, during the struvite crystallization process, many impurity ions precipitate with magnesium and phosphate, resulting in impure crystal substances, which reduces the application effect of the product as fertilizer, and therefore, the purity of struvite in the precipitated product, i.e., the specific gravity of struvite in the whole precipitated product, must be determined. The commonly used method for determining the purity of struvite comprises the following steps: 1) qualitative determination, using fourier transform infrared spectroscopy (FTIR) or X-ray diffraction (XRD) spectroscopy to compare the morphology of solid precipitates and pure struvite; 2) comparing the percentage of each element in the solid by taking the theoretical value of the pure struvite as a reference; 3) measuring the nitrogen content by using an element analysis instrument and taking the nitrogen content as a reference value; 4) quantitative X-ray diffraction method. However, these purity determination methods limit the overall production process due to the additional time required for sample preparation and analysis, since the production line needs to be halted during testing until the appropriate product purity is achieved. Therefore, the currently used methods for measuring the purity of phosphorus recovery products have the defects of complex operation, long time and the like, and the efficiency of the whole production process is greatly reduced.
It is to be noted that the information disclosed in the above background section is only for understanding the background of the present application and thus may include information that does not constitute prior art known to a person of ordinary skill in the art.
Disclosure of Invention
The main purpose of the present invention is to overcome the above mentioned drawbacks of the background art and to provide a method for soft measurement of the purity of a phosphorus recovery product.
In order to achieve the purpose, the invention adopts the following technical scheme:
a soft measurement method for the purity of a phosphorus recovery product comprises the following steps:
(1) shooting the collected phosphorus recovery product by using an optical microscope, and recording the micro-morphology structure of the phosphorus recovery product on a two-dimensional plane;
(2) processing the picture shot in the step (1) so as to measure physical parameters of the particles in the picture;
(3) measuring and calculating the following physical parameters of the particles in the picture processed in the step (2): distance nF between two points farthest from the periphery of the particlemax(ii) a Distance nF between two nearest points on the periphery of the particlemin(ii) a Elongation nEl1(ii) a Aspect ratio nEl2(ii) a Total area n of particlesA(ii) a Perimeter n of particleP(ii) a Perimeter n of particleCi(ii) a Equivalent diameter nEd(ii) a Solidity ns(ii) a Roughness nCo;
(4) Establishing a soft measurement model of the purity of the phosphorus recovery product by adopting a nonlinear partial least square NLPLS method based on the physical parameters of the product obtained by measurement and calculation in the step (3); and for a new phosphorus recovery product, performing pre-estimation analysis on the product purity by using an established NLPLS model, and updating the model by using a recursive algorithm by combining new pre-estimation analysis data with the original NLPLS model.
Further:
the processing in the step (2) comprises enhancing chromatic aberration and contrast, and performing edge tracing processing on the two-dimensional plane shape of the phosphorus recovery product in the picture.
The step (4) of establishing the soft measurement model of the purity of the phosphorus recovery product specifically comprises the following steps:
taking the measured and calculated sample morphology parameters as a data-driven sample set, and expressing as { X (i), Y (i) }, wherein X (i) expresses the ith group of input data, namely the parameters obtained in the step (3), Y (i) expresses the ith group of output data, and the input data forms a matrix X and the output data forms an output matrix Y for the purity of the product;
establishing an NLPLS model based on input and output data, which specifically comprises the following steps:
normalizing the matrixes X and Y to enable the mean value to be 0 and the variance to be 1, then expanding the input matrix in rows and columns, wherein the expansion terms are a hidden node output matrix G of a Radial Basis Function (RBF) neural network and a column vector 1 with all elements being 1, each row of G corresponds to the output of a hidden node under the action of an input vector, and the offset term coefficient of the hidden node is 1; partial Least Squares (PLS) recovery is performed on the following augmented input and output matrices:
{ [ 1X G ], Y }, the obtained NLPLS model is expressed as:
in the formula, XE represents an augmented input matrix, A and H are weight coefficient matrixes corresponding to an original input vector and an output vector of a hidden node of a RBF network respectively, b is an input offset vector, T represents transposition, c is a center vector of the hidden node, sigma is a corresponding width vector, and A and H are weight coefficient matrixes.
In step II, a hidden node center vector c, a corresponding width vector sigma, weight coefficient matrixes A and H and an input offset vector b are determined according to the following steps:
firstly, clustering input data by using a k-means clustering algorithm to obtain a hidden node center c;
secondly, adopting a p neighbor rule to calculate the width of the hidden node:
wherein N is the number of hidden node centers, and ci is the nearest p hidden node centers to the jth hidden node center;
thirdly, determining a weight coefficient matrix A, H and an offset vector b by adopting PLS regression:
and calculating a hidden node output matrix G according to the obtained hidden node center and width, then expanding the input matrix to obtain an augmented input matrix [ 1X G ], performing PLS regression on the data { [ 1X G ], Y } to obtain a PLS model parameter matrix { T, W, P, B, Q }, extracting the feature variable number a, determining the obtained parameter matrix as { Ta, Wa, Pa, Ba, Qa }, and further calculating a PLS regression coefficient matrix beta to obtain A, H and B.
In the step (4), the model is updated by a recursive algorithm by combining new pre-estimation analysis data with the original NLPLS model, and the method specifically comprises the following steps:
a. recording the input and output data of new phosphorus recovery product batches as X1 and Y1 respectively, and not containing abnormal points;
b. judging whether a new hidden node is added or not, and adding the new hidden node if the distance between the new data X1 and the center of the existing hidden node of the RBF network is greater than a set value; if the distance between the X1 and the center of the hidden node of the existing RBF network is less than or equal to a set value, the hidden node does not need to be added;
c. expanding X1, performing PLS regression to obtain an updated NLPLS model; then, the weight coefficient matrix A, H and the offset vector b are calculated according to the method of the step (c).
In step b, if the distance between the new data X1 and the existing RBF network hidden node center is larger than the set distanceValue, new hidden node center matrix is recorded as CgnewThe corresponding width vector is σgnewFor the original hidden node center matrix CgCorresponding width vector σgAnd the load matrix P is extended as follows:
if the distance between the X1 and the center of the hidden node of the existing RBF network is less than or equal to a set value, Cg、σgAnd P remains unchanged.
Extension of X1 to XE1 ═ 1X 1G 1]Where G1 is the output matrix of hidden nodes to X1, letFor data pair { XEAnd Y, performing PLS regression to obtain an updated NLPLS model.
A computer-readable storage medium storing a computer program which, when executed by a processor, performs steps (2) to (4) of the method for soft measurement of phosphorus recovery product purity.
The invention has the following beneficial effects:
the invention provides a soft measurement method for the purity of a phosphorus recovery product, which quantitatively describes a two-dimensional morphology structure of the product, and then adopts a nonlinear least squares (NLPLS) method to establish a soft measurement model between a shape factor and the product purity, so that the defects of complex operation, long time and the like of the existing purity measurement method are overcome, and the efficiency of the whole production process is greatly improved. The method is based on a microscope photographing technology, the shape factor of the product is expressed by physical parameters, a soft measurement model between the shape factor and the product purity is established by adopting a nonlinear least square method, and the model is used for predicting the product purity so as to provide operation guidance for production control and optimization. The method effectively avoids the problems of overlong time consumption, signal distortion and the like in the traditional purity measurement process, and provides a quick, simple and accurate method for measuring the purity of the phosphorus recovery product.
Drawings
FIG. 1 is a schematic flow chart of a soft measurement method for a phosphorus recovery product according to an embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be described in detail below. It should be emphasized that the following description is merely exemplary in nature and is not intended to limit the scope of the invention or its application.
It will be understood that when an element is referred to as being "secured to" or "disposed on" another element, it can be directly on the other element or be indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or be indirectly connected to the other element. In addition, the connection may be for either a fixed or coupled or communicating function.
It is to be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like are used in an orientation or positional relationship indicated in the drawings for convenience in describing the embodiments of the present invention and to simplify the description, and are not intended to indicate or imply that the referenced device or element must have a particular orientation, be constructed in a particular orientation, and be in any way limiting of the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of the present invention, "a plurality" means two or more unless specifically limited otherwise.
Referring to fig. 1, the soft measurement method for the purity of the phosphorus recovery product provided by the embodiment of the invention comprises the following steps:
when the concentration of each ion in the wastewater is Mg2+:NH4 +:PO4 3-:Ca2+When the ratio is 2:4:1:0.4, white solid precipitates such as calcium phosphate, magnesium phosphate, struvite and the like are formed. The solid precipitated products were collected, and the phosphorus recovery products were photographed using an optical microscope and their microscopic morphology on a two-dimensional plane was recorded.
And processing the shot picture, adjusting the picture background to be white, adjusting the particles to be black, enhancing the color difference and contrast, and performing edge tracing processing on the two-dimensional plane shape of the phosphorus recovery product on the picture.
The following physical parameters of the particles evident in the pictures were measured:
distance between two points farthest from the periphery of the particle (Max Feret Diameter): nFmax
Distance between the nearest two points around the particle (Min Feret Diameter): nFmin
Total Area of particles (Area): n isA
Perimeter of particle (Perimeter): n isP
The following physical quantities were calculated using the measurement parameters:
the model establishment and purity prediction process is as follows:
the method comprises the following steps: and establishing a soft measurement model of the purity of the phosphorus recovery product by adopting a nonlinear partial least squares (NLPLS) method based on the physical parameters of the product obtained by measurement and calculation. The specific method comprises the following steps:
and I, taking the measured and calculated sample morphology parameters as a data-driven sample set, and expressing as { x (i), y (i) }, wherein x (i) represents the ith group of input data, and all the parameters capable of determining the purity of the product, namely the parameters obtained in the step (3) above. y (i) represents the output data of the ith group, which is the purity of the product. Forming input data into a matrix X and forming output data into an output matrix Y;
and II, establishing an NLPLS model based on the input and output data. The method comprises the following steps:
the matrices X and Y are normalized to have a mean of 0 and a variance of 1. Then, performing row-column expansion on the input matrix, wherein expansion terms are a hidden node output matrix G of a Radial Basis Function (RBF) neural network and a column vector 1 with elements of all 1, each row of G corresponds to the output of a hidden node under the action of an input vector, and the offset term coefficient of the hidden node is 1; partial Least Squares (PLS) recovery is performed on the following augmented input and output matrices:
{ [ 1X G ], Y }, the obtained NLPLS model is expressed as:
in the formula, XE represents an augmented input matrix, A and H are weight coefficient matrixes corresponding to an original input vector and an output vector of a hidden node of the RBF network respectively, b is an input offset vector, and T represents transposition.
Unknown parameters in the NLPLS model are a hidden node center vector c, a corresponding width vector sigma, weight coefficient matrixes A and H and a model offset vector b, and the parameters are determined according to the following steps:
firstly, clustering input data by using a k-means clustering algorithm to obtain a hidden node center c; the algorithm can determine the optimal number of the clustering centers and simultaneously can ensure that the clustering centers are reasonably distributed in a data space;
secondly, adopting a p neighbor rule to calculate the width of the hidden node:
wherein N is the number of hidden node centers, and ci is the nearest p hidden node centers to the jth hidden node center.
Thirdly, determining a weight coefficient matrix A, H and an offset vector b by adopting PLS regression:
and calculating a hidden node output matrix G according to the obtained hidden node center and width, and then expanding the input matrix to obtain an augmented input matrix [ 1X G ]. And performing PLS regression on the data { [ 1X G ], Y } to obtain a PLS model parameter matrix { T, W, P, B, Q }. In order to retain all information in the subsequent model updating, the number a of extracted characteristic variables is determined by a cross-check method, the obtained parameter matrix is marked as { Ta, Wa, Pa, Ba, Qa }, and a PLS regression coefficient matrix beta is calculated by the parameter matrix and the parameter matrix, so that A, H and b are obtained.
Step two: for a new batch, performing predictive analysis on the product purity by using an established NLPLS model: and (4) transmitting the product shape parameters of the new batch to the NLPLS model, and estimating the product purity of the batch.
Step three: and when a new batch is finished, the model is updated by combining the new data with the original NLPLS model and adopting a recursion algorithm, so that the model can continuously adopt new information and adapt to the change of the process. The method comprises the following specific steps:
a. the input/output data of the newly obtained batches are X1 and Y1 (may be data of one batch or data of a plurality of accumulated batches), and do not contain outliers. Firstly, the data preprocessing is carried out on the new data by adopting the same method as the step one.
b. Judging whether a new hidden node is added:
if the distance between the new data X1 and the center of the hidden node of the existing RBF network is greater than a set value, adding a new hidden node; recording the new hidden node central matrix as CgnewThe corresponding width vector is σgnewFor the original hidden node center matrix CgCorresponding width vector σgAnd the load matrix P is extended as follows:
if the distance between the X1 and the center of the hidden node of the existing RBF network is less than or equal to a set value, the hidden node, C, does not need to be addedg、σgAnd P remains unchanged;
c. extension of X1 to XE1 ═ 1X 1G 1]Where G1 is the output matrix of hidden nodes to X1, letFor data pair { XEY } performing PLS regression to obtain a new NLPLS model:then, calculating a weight coefficient matrix A, H and a bias vector b according to the method in the third step;
d. and after obtaining a new model, returning to the step two, and using the new model for the newly obtained batch data.
The background of the present invention may contain background information related to the problem or environment of the present invention and does not necessarily describe the prior art. Accordingly, the inclusion in the background section is not an admission of prior art by the applicant.
The foregoing is a more detailed description of the invention in connection with specific/preferred embodiments and is not intended to limit the practice of the invention to those descriptions. It will be apparent to those skilled in the art that various substitutions and modifications can be made to the described embodiments without departing from the spirit of the invention, and these substitutions and modifications should be considered to fall within the scope of the invention. In the description herein, references to the description of the term "one embodiment," "some embodiments," "preferred embodiments," "an example," "a specific example," or "some examples" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Various embodiments or examples and features of various embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction. Although embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions and alterations can be made herein without departing from the scope of the claims.
Claims (8)
1. A soft measurement method for the purity of a phosphorus recovery product is characterized by comprising the following steps:
(1) shooting the collected phosphorus recovery product by using an optical microscope, and recording the micro-morphology structure of the phosphorus recovery product on a two-dimensional plane;
(2) processing the picture shot in the step (1) so as to measure physical parameters of the particles in the picture;
(3) measuring and calculating the following physical parameters of the particles in the picture processed in the step (2): distance nF between two points farthest from the periphery of the particlemax(ii) a Distance nF between two nearest points on the periphery of the particlemin(ii) a Elongation nEl1(ii) a Aspect ratio nEl2(ii) a Total area n of particlesA(ii) a Perimeter n of particleP(ii) a Perimeter n of particleCi(ii) a Equivalent diameter nEd(ii) a Solidity ns(ii) a Roughness nCo;
(4) Establishing a soft measurement model of the purity of the phosphorus recovery product by adopting a nonlinear partial least square NLPLS method based on the physical parameters of the product obtained by measurement and calculation in the step (3); and for a new phosphorus recovery product, performing pre-estimation analysis on the product purity by using an established NLPLS model, and updating the model by using a recursive algorithm by combining new pre-estimation analysis data with the original NLPLS model.
2. The method of claim 1, wherein the processing in step (2) comprises enhancing color difference and contrast to border the two-dimensional planar shape of the phosphorus recovery product in the picture.
3. The method of claim 1, wherein the step (4) of establishing a soft measurement model of the purity of the recovered phosphorus product comprises the steps of:
taking the measured and calculated sample morphology parameters as a data-driven sample set, and expressing as { X (i), Y (i) }, wherein X (i) expresses the ith group of input data, namely the parameters obtained in the step (3), Y (i) expresses the ith group of output data, and the input data forms a matrix X and the output data forms an output matrix Y for the purity of the product;
establishing an NLPLS model based on input and output data, which specifically comprises the following steps:
normalizing the matrixes X and Y to enable the mean value to be 0 and the variance to be 1, then expanding the input matrix in rows and columns, wherein the expansion terms are a hidden node output matrix G of a Radial Basis Function (RBF) neural network and a column vector 1 with all elements being 1, each row of G corresponds to the output of a hidden node under the action of an input vector, and the offset term coefficient of the hidden node is 1; partial Least Squares (PLS) recovery is performed on the following augmented input and output matrices:
{ [ 1X G ], Y }, the obtained NLPLS model is expressed as:
in the formula, XE represents an augmented input matrix, A and H are weight coefficient matrixes corresponding to an original input vector and an output vector of a hidden node of a RBF network respectively, b is an input offset vector, T represents transposition, c is a center vector of the hidden node, sigma is a corresponding width vector, and A and H are weight coefficient matrixes.
4. The method of claim 3, wherein in step ii, the center vector c of the hidden node, the corresponding width vector σ, the weight coefficient matrices a and H, and the input offset vector b are determined by:
firstly, clustering input data by using a k-means clustering algorithm to obtain a hidden node center c;
secondly, adopting a p neighbor rule to calculate the width of the hidden node:
wherein N is the number of hidden node centers, and ci is the nearest p hidden node centers to the jth hidden node center;
thirdly, determining a weight coefficient matrix A, H and an offset vector b by adopting PLS regression:
and calculating a hidden node output matrix G according to the obtained hidden node center and width, then expanding the input matrix to obtain an augmented input matrix [ 1X G ], performing PLS regression on the data { [ 1X G ], Y } to obtain a PLS model parameter matrix { T, W, P, B, Q }, extracting the feature variable number a, determining the obtained parameter matrix as { Ta, Wa, Pa, Ba, Qa }, and further calculating a PLS regression coefficient matrix beta to obtain A, H and B.
5. The method of claim 3 or 4, wherein the step (4) of updating the model with a recursive algorithm using new pre-estimated analytical data in combination with the original NLPLS model comprises the steps of:
a. recording the input and output data of new phosphorus recovery product batches as X1 and Y1 respectively, and not containing abnormal points;
b. judging whether a new hidden node is added or not, and adding the new hidden node if the distance between the new data X1 and the center of the existing hidden node of the RBF network is greater than a set value; if the distance between the X1 and the center of the hidden node of the existing RBF network is less than or equal to a set value, the hidden node does not need to be added;
c. expanding X1, performing PLS regression to obtain an updated NLPLS model; then, the weight coefficient matrix A, H and the offset vector b are calculated according to the method of the step (c).
6. The method of claim 5, wherein in step b, if the new data X1 is more than a predetermined distance from the center of hidden nodes in the existing RBF network, the new hidden node center matrix is recorded as CgnewThe corresponding width vector is σgnewFor the original hidden node center matrix CgCorresponding width vector σgAnd the load matrix P is extended as follows:
if the distance between the X1 and the center of the hidden node of the existing RBF network is less than or equal to a set value, Cg、σgAnd P remains unchanged.
8. A computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, performs steps (2) to (4) of the method for soft measurement of phosphorus recovery product purity of any of claims 1 to 7.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110366780.1A CN113096087A (en) | 2021-04-06 | 2021-04-06 | Soft measurement method for purity of phosphorus recovery product |
PCT/CN2022/085201 WO2022213958A1 (en) | 2021-04-06 | 2022-04-03 | Soft measurement method for purity of phosphorus recovery product |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110366780.1A CN113096087A (en) | 2021-04-06 | 2021-04-06 | Soft measurement method for purity of phosphorus recovery product |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113096087A true CN113096087A (en) | 2021-07-09 |
Family
ID=76674043
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110366780.1A Pending CN113096087A (en) | 2021-04-06 | 2021-04-06 | Soft measurement method for purity of phosphorus recovery product |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN113096087A (en) |
WO (1) | WO2022213958A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022213958A1 (en) * | 2021-04-06 | 2022-10-13 | 清华大学深圳国际研究生院 | Soft measurement method for purity of phosphorus recovery product |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101872182A (en) * | 2010-05-21 | 2010-10-27 | 杭州电子科技大学 | Batch process monitoring method based on recursive non-linear partial least square |
CN101872387A (en) * | 2010-05-21 | 2010-10-27 | 杭州电子科技大学 | Soft-sensing method for beer flavor |
CN109522514A (en) * | 2018-11-07 | 2019-03-26 | 浙江工业大学之江学院 | The PX adsorbing separation purity soft-measuring modeling method corrected based on time difference Recursive PLS and adaptive model |
CN110357676A (en) * | 2019-05-30 | 2019-10-22 | 中山大学 | The method for determining guanite recycling nitrogen phosphorus technological parameter based on Response Surface Method |
CN111192236A (en) * | 2019-12-13 | 2020-05-22 | 沈阳化工大学 | SVI (singular value index) measurement method based on activated sludge phase difference microscopic image |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4814858A (en) * | 1988-04-06 | 1989-03-21 | Hitachi, Ltd. | Method of measuring color purity of color display and apparatus therefor |
CN113096087A (en) * | 2021-04-06 | 2021-07-09 | 清华大学深圳国际研究生院 | Soft measurement method for purity of phosphorus recovery product |
-
2021
- 2021-04-06 CN CN202110366780.1A patent/CN113096087A/en active Pending
-
2022
- 2022-04-03 WO PCT/CN2022/085201 patent/WO2022213958A1/en active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101872182A (en) * | 2010-05-21 | 2010-10-27 | 杭州电子科技大学 | Batch process monitoring method based on recursive non-linear partial least square |
CN101872387A (en) * | 2010-05-21 | 2010-10-27 | 杭州电子科技大学 | Soft-sensing method for beer flavor |
CN109522514A (en) * | 2018-11-07 | 2019-03-26 | 浙江工业大学之江学院 | The PX adsorbing separation purity soft-measuring modeling method corrected based on time difference Recursive PLS and adaptive model |
CN110357676A (en) * | 2019-05-30 | 2019-10-22 | 中山大学 | The method for determining guanite recycling nitrogen phosphorus technological parameter based on Response Surface Method |
CN111192236A (en) * | 2019-12-13 | 2020-05-22 | 沈阳化工大学 | SVI (singular value index) measurement method based on activated sludge phase difference microscopic image |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022213958A1 (en) * | 2021-04-06 | 2022-10-13 | 清华大学深圳国际研究生院 | Soft measurement method for purity of phosphorus recovery product |
Also Published As
Publication number | Publication date |
---|---|
WO2022213958A1 (en) | 2022-10-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116186634B (en) | Intelligent management system for construction data of building engineering | |
Sun et al. | Automatic remotely sensed image classification in a grid environment based on the maximum likelihood method | |
Leighty et al. | Dynamic metabolic flux analysis (DMFA): a framework for determining fluxes at metabolic non-steady state | |
Blangero et al. | A kernel of truth: statistical advances in polygenic variance component models for complex human pedigrees | |
Yan et al. | Large-scale crop mapping from multi-source optical satellite imageries using machine learning with discrete grids | |
Sun et al. | Retrieval of rapeseed leaf area index using the PROSAIL model with canopy coverage derived from UAV images as a correction parameter | |
CN111612055B (en) | Weather situation typing method, air pollution condition prediction method and device | |
Wang et al. | SSRNet: In-field counting wheat ears using multi-stage convolutional neural network | |
Li et al. | A multi-scale cucumber disease detection method in natural scenes based on YOLOv5 | |
CN111126511B (en) | Vegetation index fusion-based LAI quantitative model establishment method | |
CN107860889A (en) | The Forecasting Methodology and equipment of the soil organism | |
WO2022213958A1 (en) | Soft measurement method for purity of phosphorus recovery product | |
Sun et al. | Wheat head counting in the wild by an augmented feature pyramid networks-based convolutional neural network | |
CN112906300A (en) | Polarized SAR (synthetic Aperture Radar) soil humidity inversion method based on two-channel convolutional neural network | |
CN112735536A (en) | Single cell integrated clustering method based on subspace randomization | |
CN114758170B (en) | Three-branch three-attention mechanism hyperspectral image classification method combined with D3D | |
Chen et al. | Improving the practicability of remote sensing data-assimilation-based crop yield estimations over a large area using a spatial assimilation algorithm and ensemble assimilation strategies | |
Yu et al. | Combining the critical nitrogen concentration and machine learning algorithms to estimate nitrogen deficiency in rice from UAV hyperspectral data | |
Feng et al. | Merging and downscaling soil moisture data from CMIP6 projections using deep learning method | |
CN111612734B (en) | Background clutter characterization method based on image structure complexity | |
CN110956178B (en) | Plant growth measuring method and system based on image similarity calculation and electronic equipment | |
CN116561498B (en) | Global forest carbon sequestration potential high-precision calculation method and system | |
CN109800690B (en) | Nonlinear hyperspectral image mixed pixel decomposition method and device | |
CN112330694A (en) | Plant wilting degree calculation method and system | |
Xu et al. | Segmentation and counting of wheat spike grains based on deep learning and textural feature |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210709 |
|
RJ01 | Rejection of invention patent application after publication |