CN112734166B - Copper industry data robust coordination and significant error detection method - Google Patents
Copper industry data robust coordination and significant error detection method Download PDFInfo
- Publication number
- CN112734166B CN112734166B CN202011512638.5A CN202011512638A CN112734166B CN 112734166 B CN112734166 B CN 112734166B CN 202011512638 A CN202011512638 A CN 202011512638A CN 112734166 B CN112734166 B CN 112734166B
- Authority
- CN
- China
- Prior art keywords
- variable
- data
- coordination
- lower limit
- limit
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 31
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 title claims abstract description 19
- 229910052802 copper Inorganic materials 0.000 title claims abstract description 19
- 239000010949 copper Substances 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 70
- 239000000463 material Substances 0.000 claims abstract description 49
- 238000011084 recovery Methods 0.000 claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000004519 manufacturing process Methods 0.000 claims abstract description 20
- 238000005457 optimization Methods 0.000 claims abstract description 17
- 238000003723 Smelting Methods 0.000 claims abstract description 6
- 238000006243 chemical reaction Methods 0.000 claims abstract description 5
- 230000008569 process Effects 0.000 claims description 23
- 239000002184 metal Substances 0.000 claims description 17
- 229910052751 metal Inorganic materials 0.000 claims description 17
- 230000035772 mutation Effects 0.000 claims description 7
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 6
- 210000000349 chromosome Anatomy 0.000 claims description 6
- 230000008878 coupling Effects 0.000 claims description 6
- 238000010168 coupling process Methods 0.000 claims description 6
- 238000005859 coupling reaction Methods 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 108090000623 proteins and genes Proteins 0.000 claims description 6
- 238000009795 derivation Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 claims description 3
- 230000014509 gene expression Effects 0.000 claims description 3
- 239000004615 ingredient Substances 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims 1
- 238000004364 calculation method Methods 0.000 abstract description 7
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 abstract description 2
- NINIDFKCEFEMDL-UHFFFAOYSA-N Sulfur Chemical compound [S] NINIDFKCEFEMDL-UHFFFAOYSA-N 0.000 abstract description 2
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 abstract description 2
- 229910052737 gold Inorganic materials 0.000 abstract description 2
- 239000010931 gold Substances 0.000 abstract description 2
- 229910052717 sulfur Inorganic materials 0.000 abstract description 2
- 239000011593 sulfur Substances 0.000 abstract description 2
- 239000012141 concentrate Substances 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 4
- 238000011426 transformation method Methods 0.000 description 4
- 238000012733 comparative method Methods 0.000 description 3
- 239000002893 slag Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229910052500 inorganic mineral Inorganic materials 0.000 description 2
- 239000011707 mineral Substances 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 238000010583 slow cooling Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000007664 blowing Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 239000002904 solvent Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/067—Enterprise or organisation modelling
-
- 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
- G06F17/15—Correlation function computation including computation of convolution operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
Abstract
The invention provides a robust coordination and significant error detection method for copper industry data, which relates to the technical field of information, and adopts real material inventory data, and firstly, the inventory data standard deviation of each data item is roughly calculated according to the inventory data; according to the recovery rate of copper, sulfur, gold and silver metal in a smelting plant of a copper production enterprise; in order to prevent the interference of significant errors in inventory data to other data items during data coordination, the invention establishes a robust data coordination model; in order to realize rapid data coordination and significant error detection of all material properties for inventory data of materials in a smelting plant, the invention adopts a Lagrange multiplier conversion method, and a high-dimensional variable space is represented by a low-dimensional constraint multiplier space. And in addition, adopting a differential evolution algorithm to carry out optimization iterative solution. The method has the advantages that the obtained result is more accurate in remarkable error detection, the coordinated data is more stable, the calculation efficiency meets the actual requirements, and the method can be popularized and applied in copper production enterprises and other production enterprises.
Description
Technical Field
The invention relates to the technical field of information, in particular to the technologies of robust calculation, meta-heuristic optimization and the like, and discloses a method for robust coordination and significant detection of colored copper production data by combining a robust calculation method with a meta-heuristic method.
Background
The smelting factory is taken as an important component part of the whole nonferrous copper production, and the recovery condition of valuable metals such as copper, sulfur, gold, silver and the like directly affects the production control and profit accounting of the whole enterprise to play a decisive role. The metal balance report of the month can be obtained through measuring the quality of all stock materials at the end of the month and testing the grades of valuable elements contained in the stock materials, and the production condition of the month can be measured through calculating the recovery rate and the unknown loss rate of all the valuable elements. Under the normal production condition, the element recovery rate is in the artificial experience estimation interval, but is influenced by error factors such as material metering, sampling, inspection and analysis and the like, so that the unknown loss of metal balance caused by measurement errors and obvious errors is large, and the element mass conservation deviation is large. Therefore, significant error detection and data coordination are required to be performed on the checked data, so that the influence of errors on metal balance is reduced.
With the wide development of research application work, the data coordination of a dynamic system can be realized by using a meta-heuristic method and a gradient optimization representative method. The methods all assume that errors in acquired data conform to Gaussian distribution, however, significant errors exist when the data of industrial inventory are affected by manual operation, sampling and the like. Therefore, direct data coordination of the data can spread significant error differences in the data to other data items to cause the coordinated result to deviate from a normal value, and therefore, a distribution detection method is adopted to detect the significant errors of the data.
In view of this, in addition to the significant error detection method based on measurement errors and measurement residuals, a robust objective function may be used to limit the contribution of data items with residuals larger than tuning parameters in the objective function (Hou Wenting. Data driven robust optimization based power system multisource coordinated optimization scheduling study [ D ]. University of guangxi, 2019.). However, because the dimension of the data coordination data item of the copper production enterprise is high, and the constraint condition of the data coordination model is a bilinear form of the product between variable attributes, which is difficult to decouple, the algorithm iteration is slow and difficult to converge due to the fact that the solution is directly carried out by adopting a meta heuristic method; this time, because there may be multiple significant error terms at the same time, a cyclic detection method is required, which may result in an excessively long detection time. If a robust data coordination model is used, although multiple significant errors can be detected simultaneously, the objective function becomes nonlinear, resulting in difficulty in solving the model. If the idea of variable conversion is adopted, the original variable with high dimension is represented by a constraint multiplier with lower dimension, and a plurality of significant error items are determined simultaneously by adopting a robust data coordination technology, the significant error in the data is expected to be determined simultaneously when the solution is rapidly solved.
Disclosure of Invention
The invention aims to solve the technical problem of overcoming the defects existing in the prior art and providing a robust coordination and significant error detection method for copper industry data.
The invention is realized by the following technical scheme: the robust coordination and significant error detection method for the copper industry data is characterized by comprising the following steps:
S1: data preprocessing
Reading current month inventory data comprising the quality of materials and grade values of corresponding valuable elements by a production site SAP database, calculating the mean value and standard deviation of the measured values, and estimating the recovery rate and the nameless loss rate of each metal element of the smelting plant according to the current month production condition;
S2: data robust coordinated modeling
Firstly, considering stable recovery rate of metal quantity and unknown loss rate of a factory;
S21: the expressions of the metal amount recovery rate and the nameless loss rate of the metal element are respectively shown in the following formulas (1) (2):
Wherein x represents the mass of the material, ω represents the grade value of the material element, subscript in represents the variable attribute as input or initial end of life, subscript end represents the variable attribute as end of life, subscript out represents the variable attribute as output, subscript loss represents the material attribute as famous loss, α ul (i) represents the nameless loss rate constraint target of element i, and α re (i) represents the recovery rate constraint target of element i;
s22: the robust function based on design can be used for establishing a data coordination model which is shown in the formula (3) and takes the robust function as an objective function, and the recovery rate of each valuable element, the steady rate of unknown loss and the physical upper and lower limits of each variable form a constraint set by deduction of element conservation:
In order to eliminate the influence of different variable orders on the coordination result, the coordinated relative residual is expressed by dividing the coordinated residual by the relative standard deviation;
ρ Welsch is represented as (4):
Wherein the method comprises the steps of Q x (j)＝σx (j)·xm (j),sw is tuning parameter, sigma x (j) is relative standard deviation of j-th material,/>Wherein/>For the relative standard deviation of the ith element corresponding to the jth material, g ul (i) and g re (i) are respectively the nameless loss rate and recovery rate constraint derivation type,/>X l (j) and x u (j) correspond to the lower and upper coordination limits of the material quality, and w l (i)(j) and w u (i)(j) are the lower and upper coordination limits of the ith grade of the jth material;
S3: multiplier dimension reduction conversion of high-dimension variable
The Lagrangian function corresponding to the original model when considering only the equality constraint is as shown in (5):
wherein lambda ul (i)、λre (i) is the Lagrangian multiplier corresponding to the ith element recovery constraint and the unknown loss rate constraint;
s31: for L And/>The simultaneous solving can be achieved:
Wherein lambertw (0, x) is Lambert W Function, representing that the number of bits in the definition field (0, ++ infinity), s w is the tuning constant of the Welsch function. Lambda (i)＝-λul (i)-λre (i) when the material attribute is input or initial balance, lambda (i)＝λul (i)+λre (i) when the material attribute is final balance; lambda (i)＝λul (i)+λre (i)·kre (i) when the material attribute is output; when the material attribute is a named loss, lambda (i)＝λul (i)·kul (i);
S32: and obtaining the upper and lower coordination limits of each coordination variable according to the historical data of the inventory, the results of the multiple measurements and the physical limit of the production workshop. To further reduce the dimensions of the multiplier subspace, the variables are segmented during the reconciliation process. From equation (6), it can be seen that a uniquely determined set of variables can be solved for each set of multipliers λ (i) AndAnd only the quantity of the same material and the element grade have a coupling relation;
Under the condition of considering the upper and lower limits of the variables, each is estimated first The variables are only considered variables/>Range of variation under upper and lower limit limits:
according to positive and negative conditions of lambda (i), the estimated interval and the set interval of the variable can be classified into the following three states:
state ①: if the range is completely beyond the upper and lower limit constraint range of the variable, the solution of the variable which is not limited by the upper and lower limits is outside the upper and lower limit range, the variable is directly taken as the upper limit w l (i)(j) (if lambda (i) is less than 0) or the lower limit w l (i)(j) (if lambda (i) is more than 0) according to the lambda (i) sign;
state ②: if the range is completely contained in the upper and lower limits of the variable, the upper and lower limits of the variable are not required to be concerned in the solving process, and the solution of the variable is necessarily within the upper and lower limits;
State ③: if the range is intersected with the upper limit and the lower limit, and no inclusion relation exists, marking is carried out, and the upper limit and the lower limit of the variable still need to be concerned in the solving process.
Solving after estimation, directly fixing the variables in the state ① to an upper limit or a lower limit, wherein the variables in the state ②、③ are not constrained by the upper limit and the lower limit, and directly obtaining a solution without performing next operation if the group of variables are in the upper limit and the lower limit after solving; if the variable exceeds the upper and lower limit, the next search operation on the solution is executed, and the search range is according to the variableThe cases of (2) are classified into the following three cases:
1) Variable(s) Fixed to the upper limit, the variable/>The decoupling relationship can be directly solved according to the estimated state, wherein if the variable of the state ③ exceeds the upper limit or the lower limit, the variable is pulled back to the upper limit or the lower limit.
2) Variable(s)Fixed to the lower limit, variable/>The decoupling relationship can be directly solved according to the estimated state, wherein if the variable of the state ③ exceeds the upper limit or the lower limit, the variable is pulled back to the upper limit or the lower limit.
3) Variable(s)Without touching the upper and lower limits, the variable/>There is still a coupling relationship between each/>There are two cases of touching the upper and lower limits, and no touching the upper and lower limits (the variable of state ① only has the case of touching the upper and lower limits, and the variable of state ② only has the case of not touching the upper and lower limits), wherein the variable of state ③ is pulled back to the upper or lower limit if it exceeds the upper or lower limit.
S4: long-term ingredient planning global optimization:
The basic idea of the DE algorithm is to simulate a random model of biological evolution, such that those more environmentally adapted individuals are preserved by iterative substitution (mutation, crossover and selection). In the evolution process, PSO algorithm characteristics are combined, global optimal individual guide information is added, whether the algorithm converges or not is considered by taking turns as a unit for judging population density in calculation, and the random search performance of the algorithm is increased, so that the algorithm can jump out of a local optimal solution more easily. The population is initialized to [ P 1,P2…Ppsize ], where each individual P contains a set of Lagrange multiplier variables, psize is the population size. The population density is defined as a density threshold TI of the population, a continuous non-convergence round threshold TNC and an fitness function threshold TF shown in a formula (8). The fitness function of DE in performing the solution search is as in equation (9). The variation process of the algorithm is shown as a formula (10), the crossover process is shown as a formula (11), and the selection process is shown as a formula (12).
Wherein the method comprises the steps ofFor two different individuals selected randomly in the population except for variant individuals, V i is the new individual generated, G max is the maximum number of iterations, G is the current number of iterations F 0 is the evolution factor in interval [0,2 ].
Where CR is crossover probability, j r is a random integer of [ 1.. psize ], U j,i is the j gene of the i chromosome after crossover, and V j,i is the j gene of the i chromosome after mutation.
The following is a solution step for a robust data coordination model by using a DE method in combination with a multiplier transformation technique:
s41: initializing an overall programming vector population Wherein each individual comprises a set of lagrangian multipliers, d being the population size;
S42: performing individual mutation operation by adopting a formula (10):
s43: performing crossover operation by adopting a formula (11):
s44: obtaining new population after the above process And calculates a corresponding fitness function (9);
s45: mixing the new population and the old population, reserving the first d individuals with longer total duration of the batching scheme in a greedy optimization mode to form a new population, and marking the optimal individuals as Gbest;
s46: calculating population density by adopting a formula (8), if the density is smaller than a preset TF, retaining Gbest and reinitializing the rest population individuals;
if the algorithm does not reach the maximum iteration number, returning to the step S42; otherwise, returning the optimal individual and the corresponding data coordination value.
S5: and (3) significant error detection:
The relative residual e i of the data reconciliation is calculated and the relationship of the residual to the Welsch tuning parameter s is determined, and marked as a significant error if e i > s.
The beneficial effects of the invention are as follows: the data robust coordination and significant error detection method combines the multiplier transformation method, the robust optimization method and the meta heuristic optimization method, solves the problem that the traditional direct data coordination is affected by significant error items, and can solve the problem of difficult solution to the high-dimensional data coordination problem containing complex bilinear constraints. The robust data coordination method can realize data coordination and significant error detection of the data item at the same time, and reduces the cycle detection problem caused by adopting distributed detection. . In addition, the variable dimension of the meta-heuristic algorithm is greatly reduced by the multiplier transformation method, the search space of the algorithm is reduced, the efficiency of iterative optimization is increased, and the calculation efficiency of the method meets the actual application requirements.
Drawings
FIG. 1 is a schematic diagram of a flow of a production material from a nonferrous copper manufacturing enterprise;
FIG. 2 is a flow chart of the application of the present invention;
FIG. 3 is a box plot of relative variation indicators for the method of the present invention and the comparative method;
FIG. 4 is a diagram showing a comparison of AVTI statistical parameters for the inventive method and the comparative method;
Figure 5 is a graph of OP statistical parameters for the inventive method and the comparative method.
Detailed Description
The present invention will be described in further detail below with reference to the drawings and preferred embodiments, so that those skilled in the art can better understand the technical solutions of the present invention.
As shown in the figure, the invention provides a method for robust coordination and significant error detection of copper industry data, which is characterized by comprising the following steps:
S1: data preprocessing
Reading current month inventory data comprising the quality of materials and grade values of corresponding valuable elements by a production site SAP database, calculating the mean value and standard deviation of the measured values, and estimating the recovery rate and the nameless loss rate of each metal element of the smelting plant according to the current month production condition;
S2: data robust coordinated modeling
Firstly, considering stable recovery rate of metal quantity and unknown loss rate of a factory;
S21: the expressions of the metal amount recovery rate and the nameless loss rate of the metal element are respectively shown in the following formulas (1) (2):
Wherein x represents the mass of the material, ω represents the grade value of the material element, subscript in represents the variable attribute as input or initial end of life, subscript end represents the variable attribute as end of life, subscript out represents the variable attribute as output, subscript loss represents the material attribute as famous loss, α ul (i) represents the nameless loss rate constraint target of element i, and α re (i) represents the recovery rate constraint target of element i;
s22: the robust function based on design can be used for establishing a data coordination model which is shown in the formula (3) and takes the robust function as an objective function, and the recovery rate of each valuable element, the steady rate of unknown loss and the physical upper and lower limits of each variable form a constraint set by deduction of element conservation:
In order to eliminate the influence of different variable orders on the coordination result, the coordinated relative residual is expressed by dividing the coordinated residual by the relative standard deviation;
ρ Welsch is represented as (4):
Wherein the method comprises the steps of Q x (j)＝σx (j)·xm (j),sw is tuning parameter, sigma x (j) is relative standard deviation of j-th material,/>Wherein/>For the relative standard deviation of the ith element corresponding to the jth material, g ul (i) and g re (i) are respectively the nameless loss rate and recovery rate constraint derivation type,/>X l (j) and x u (j) correspond to the lower and upper coordination limits of the material quality, and w l (i)(j) and w u (i)(j) are the lower and upper coordination limits of the ith grade of the jth material;
S3: multiplier dimension reduction conversion of high-dimension variable
The Lagrangian function corresponding to the original model when considering only the equality constraint is as shown in (5):
wherein lambda ul (i)、λre (i) is the Lagrangian multiplier corresponding to the ith element recovery constraint and the unknown loss rate constraint;
s31: for L And/>The simultaneous solving can be achieved:
Wherein lambertw (0, x) is Lambert W Function, representing that the number of bits in the definition field (0, ++ infinity), s w is the tuning constant of the Welsch function. Lambda (i)＝-λul (i)-λre (i) when the material attribute is input or initial balance, lambda (i)＝λul (i)+λre (i) when the material attribute is final balance; lambda (i)＝λul (i)+λre (i)·kre (i) when the material attribute is output; when the material attribute is a named loss, lambda (i)＝λul (i)·kul (i);
S32: and obtaining the upper and lower coordination limits of each coordination variable according to the historical data of the inventory, the results of the multiple measurements and the physical limit of the production workshop. To further reduce the dimensions of the multiplier subspace, the variables are segmented during the reconciliation process. From equation (6), it can be seen that a uniquely determined set of variables can be solved for each set of multipliers λ (i) AndAnd only the quantity of the same material and the element grade have a coupling relation;
Under the condition of considering the upper and lower limits of the variables, each is estimated first The variables are only considered variables/>Range of variation under upper and lower limit limits:
according to positive and negative conditions of lambda (i), the estimated interval and the set interval of the variable can be classified into the following three states:
state ①: if the range is completely beyond the upper and lower limit constraint range of the variable, the solution of the variable which is not limited by the upper and lower limits is outside the upper and lower limit range, the variable is directly taken as the upper limit w l (i)(j) (if lambda (i) is less than 0) or the lower limit w l (i)(j) (if lambda (i) is more than 0) according to the lambda (i) sign;
state ②: if the range is completely contained in the upper and lower limits of the variable, the upper and lower limits of the variable are not required to be concerned in the solving process, and the solution of the variable is necessarily within the upper and lower limits;
State ③: if the range is intersected with the upper limit and the lower limit, and no inclusion relation exists, marking is carried out, and the upper limit and the lower limit of the variable still need to be concerned in the solving process.
Solving after estimation, directly fixing the variables in the state ① to an upper limit or a lower limit, wherein the variables in the state ②、③ are not constrained by the upper limit and the lower limit, and directly obtaining a solution without performing next operation if the group of variables are in the upper limit and the lower limit after solving; if the variable exceeds the upper and lower limit, the next search operation on the solution is executed, and the search range is according to the variableThe cases of (2) are classified into the following three cases:
1) Variable(s) Fixed to the upper limit, the variable/>The decoupling relationship can be directly solved according to the estimated state, wherein if the variable of the state ③ exceeds the upper limit or the lower limit, the variable is pulled back to the upper limit or the lower limit.
2) Variable(s)Fixed to the lower limit, variable/>The decoupling relationship can be directly solved according to the estimated state, wherein if the variable of the state ③ exceeds the upper limit or the lower limit, the variable is pulled back to the upper limit or the lower limit.
3) Variable(s)Without touching the upper and lower limits, the variable/>There is still a coupling relationship between each/>There are two cases of touching the upper and lower limits, and no touching the upper and lower limits (the variable of state ① only has the case of touching the upper and lower limits, and the variable of state ② only has the case of not touching the upper and lower limits), wherein the variable of state ③ is pulled back to the upper or lower limit if it exceeds the upper or lower limit.
S4: long-term ingredient planning global optimization:
The basic idea of the DE algorithm is to simulate a random model of biological evolution, such that those more environmentally adapted individuals are preserved by iterative substitution (mutation, crossover and selection). In the evolution process, PSO algorithm characteristics are combined, global optimal individual guide information is added, whether the algorithm converges or not is considered by taking turns as a unit for judging population density in calculation, and the random search performance of the algorithm is increased, so that the algorithm can jump out of a local optimal solution more easily. The population is initialized to [ P 1,P2…Ppsize ], where each individual P contains a set of Lagrange multiplier variables, psize is the population size. The population density is defined as a density threshold TI of the population, a continuous non-convergence round threshold TNC and an fitness function threshold TF shown in a formula (8). The fitness function of DE in performing the solution search is as in equation (9). The variation process of the algorithm is shown as a formula (10), the crossover process is shown as a formula (11), and the selection process is shown as a formula (12).
Wherein the method comprises the steps ofFor two different individuals selected randomly in the population except for variant individuals, V i is the new individual generated, G max is the maximum number of iterations, G is the current number of iterations F 0 is the evolution factor in interval [0,2 ].
Where CR is crossover probability, j r is a random integer of [ 1.. psize ], U j,i is the j gene of the i chromosome after crossover, and V j,i is the j gene of the i chromosome after mutation.
The following is a solution step for a robust data coordination model by using a DE method in combination with a multiplier transformation technique:
s41: initializing an overall programming vector population Wherein each individual comprises a set of lagrangian multipliers, d being the population size;
S42: performing individual mutation operation by adopting a formula (10):
s43: performing crossover operation by adopting a formula (11):
s44: obtaining new population after the above process And calculates a corresponding fitness function (9);
s45: mixing the new population and the old population, reserving the first d individuals with longer total duration of the batching scheme in a greedy optimization mode to form a new population, and marking the optimal individuals as Gbest;
s46: calculating population density by adopting a formula (8), if the density is smaller than a preset TF, retaining Gbest and reinitializing the rest population individuals;
if the algorithm does not reach the maximum iteration number, returning to the step S42; otherwise, returning the optimal individual and the corresponding data coordination value.
S5: and (3) significant error detection:
The relative residual e i of the data reconciliation is calculated and the relationship of the residual to the Welsch tuning parameter s is determined, and marked as a significant error if e i > s.
Tuning parameters s of the robust model are set to 2.9846, and upper and lower limits of the material storage bin are set as shown in table 1. In the aspect of data coordination, the method and the data coordination method using PSO as a solver are respectively adopted to carry out comparison experiments, the iteration times (Iterations, IT) of the two methods, the fitness function values (Fitness function value, FF), the constraint residuals (Constrained residuals, CR) and the relative variation indexes of the data coordination results are respectively counted, wherein the relative variation index formula is shown as (13). And respectively manual methods, adopting a PSO-based data coordination method and adopting CV values of each inventory data item of the method. Table 2 shows the results of data reconciliation experiments for 1 to 6 months, and FIG. 3 shows a graph comparing CV value boxes of the method of the present invention with those of PSO-based data reconciliation method. In the aspect of significant error detection, the detection method is adopted to carry out comparison experiments with detection methods based on a residual error detection method, logistics robust functions and Fair robust functions, and the detection accuracy (OP) and Type 1 error Average Value (AVTI) of the methods are respectively counted. Fig. 4 and 5 are AVTI and OP parameter comparison graphs of the above method. The comprehensive chart results show that the method is superior to the PSO-based data coordination method in precision performance and operation efficiency, and the instability of inventory data can be greatly reduced, so that the coordinated data is more approximate to a true value.
TABLE 1 upper and lower limits of material storage bin
| Bin name | Material | Lower limit of storage | Upper limit of inventory | Unit (B) |
| Copper matte bin | Matte copper | 0 | 7000 | t |
| Concentrate bin | Concentrate | 0 | 1400 | m3 |
| Proportioning bin | Solvent, concentrate | 0 | 4400 | t |
| Dewatering system | Matte copper | 0 | 2000 | t |
| Slag concentrate warehouse | Slag concentrate | 0 | 1400 | mw |
| Mineral separation slow cooling storage yard | Top blowing furnace reducing slag | 0 | 40000 | t |
| Mineral separation slow cooling storage yard | Beneficiation returns high copper material | 0 | 5000 | t |
Table 2 1-6 month data coordination experiment
The beneficial effects of the invention are as follows: the data robust coordination and significant error detection method combines the multiplier transformation method, the robust optimization method and the meta heuristic optimization method, solves the problem that the traditional direct data coordination is affected by significant error items, and can solve the problem of difficult solution to the high-dimensional data coordination problem containing complex bilinear constraints. The robust data coordination method can realize data coordination and significant error detection of the data item at the same time, and reduces the cycle detection problem caused by adopting distributed detection. . In addition, the variable dimension of the meta-heuristic algorithm is greatly reduced by the multiplier transformation method, the search space of the algorithm is reduced, the efficiency of iterative optimization is increased, and the calculation efficiency of the method meets the actual application requirements.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (1)
1. The robust coordination and significant error detection method for the copper industry data is characterized by comprising the following steps:
s1: data preprocessing:
Reading current month inventory data comprising the quality of materials and grade values of corresponding valuable elements by a production site SAP database, calculating the mean value and standard deviation of the measured values, and estimating the recovery rate and the nameless loss rate of each metal element of the smelting plant according to the current month production condition;
s2: data robust coordination modeling:
firstly, considering stable recovery rate of metal quantity and unknown loss rate of a factory;
s21: the expressions of the metal amount recovery rate and the nameless loss rate of the metal element are respectively shown in the following formulas (1) (2):
Wherein the method comprises the steps of Representing the mass of the material,/>Representing the grade value of the material element, subscript/>Representing the variable attribute as investment or initial balance, subscript/>Representing the variable attribute as the end-of-term balance, subscript/>Representing variable attribute as yield, subscript/>Representing the material attribute as a famous loss,/>Representing element/>Target of unknown loss rate constraint,/>Representing element/>Is a recovery constraint goal of (2);
S22: the robust function based on design can build a data coordination model which is shown in the formula (3) and takes the robust function as an objective function for data coordination of metal balance, and the recovery rate of each valuable element, the steady unknown loss rate and the physical upper and lower limits of each variable form a constraint set are obtained through derivation of element conservation:
in order to eliminate the influence of different variable orders on the coordination result, the coordinated relative residual is expressed by dividing the coordinated residual by the relative standard deviation;
The formula is shown as (4):
wherein/> ，For tuning parameters,/>For/>Relative standard deviation of individual materials,/>Wherein/>To correspond to the/>Item number/>Relative standard deviation of individual elements,/>And/>Derived from the unknown loss rate and recovery rate constraints, respectively,/>，/>，/>And/>A lower coordination limit and an upper coordination limit corresponding to the material quality/>And/>Is the firstSeed material/>A lower coordination limit and an upper coordination limit of seed grades;
s3: multiplier dimension reduction conversion of high-dimension variables:
The Lagrangian function corresponding to the original model when considering only the equality constraint is as shown in (5):
Wherein the method comprises the steps of 、/>To correspond to the/>Lagrangian multipliers with seed element recovery constraints and nameless loss rate constraints;
s31: for L And/>The simultaneous solving can be achieved:
wherein/> Lambert W Function, indicated in the definition field (0, + -infinity) on a single-valued function,/(I)Tuning constant for Welsch function, material property is when put into or at the beginning of the periodWhen the material attribute is the end-of-term balance,/>; When the property of the material is yield,/>; When the material attribute is a loss of name,/>；
S32: based on the history data of inventory, the result of multiple measurements and the physical limitation of the production plant, the upper and lower coordination limits of each coordination variable can be obtained, and in order to further reduce the dimension of the multiplier space, each variable is processed in a sectioning manner in the coordination process, and the formula (6) shows that for each group of multipliersCan solve for a uniquely determined set of variables/>/>And only the quantity of the same material and the element grade have a coupling relation;
Under the condition of considering the upper and lower limits of the variables, each is estimated first The variables are only considered variables/>Range of variation under upper and lower limit limits: /(I)According to/>The pre-estimated interval of the variable and the variable setting interval can be classified into the following three states:
State ①: if the range is completely beyond the upper and lower limit constraint range of the variable, the solution without the upper and lower limit constraint is not limited to fall outside the upper and lower limit range, according to The sign directly takes the variable, if/>The < 0 variable is its upper limit/>Or if/>The variable > 0 is its lower limit/>；
State ②: if the range is completely contained in the upper and lower limits of the variable, the upper and lower limits of the variable are not required to be concerned in the solving process, and the solution of the variable is necessarily within the upper and lower limits;
State ③: if the range is intersected with the upper limit range and the lower limit range, marking is carried out, and the upper limit and the lower limit of the variables still need to be concerned in the solving process;
Solving after estimation, directly fixing the variables in the state ① to an upper limit or a lower limit, wherein the variables in the state ②、③ are not constrained by the upper limit and the lower limit, and directly obtaining a solution without performing next operation if the group of variables are in the upper limit and the lower limit after solving; if the variable exceeds the upper and lower limit, the next search operation on the solution is executed, and the search range is according to the variable The cases of (2) are classified into the following three cases:
1) Variable(s) Fixed to the upper limit, the variable/>The decoupling relation can be directly solved according to the estimated state, wherein if the variable of the state ③ exceeds the upper limit or the lower limit, the variable is pulled back to the upper limit or the lower limit;
2) Variable(s) Fixed to the lower limit, variable/>The decoupling relation can be directly solved according to the estimated state, wherein if the variable of the state ③ exceeds the upper limit or the lower limit, the variable is pulled back to the upper limit or the lower limit;
3) Variable(s) Without touching the upper and lower limits, the variable/>There is still a coupling relationship between each/>There are two cases where the upper and lower limits are reached and not reached, wherein the variable of state ③ is pulled back to the upper or lower limit if it exceeds the upper or lower limit;
S4: long-term ingredient planning global optimization:
Initializing a population [ P1, P2 … Ppsize ], wherein each individual P comprises a group of Lagrange multiplier variables, psize is the size of the population, the population density is defined as shown in a formula (8), the density threshold TI of the population, the continuous non-convergence round threshold TNC and the fitness function threshold TF are defined, the fitness function of DE in solving searching is shown in a formula (9), the variation process of an algorithm is shown in a formula (10), the crossover process is shown in a formula (11), and the selection process is shown in a formula (12):
Wherein the method comprises the steps of ，/>For two different individuals in the population selected randomly, except for variant individuals,/>For the new individuals generated,/>For maximum iteration number,/>For the current iteration number/>Is an evolution factor in interval [0,2 ];
Wherein the method comprises the steps of For cross probability,/>Random integer of [ 1.. psize ], v >For/>The first chromosome after crossingIndividual genes,/>For/>Post-mutation chromosome (s)/>A gene;
The following is a solution step for a robust data coordination model by using a DE method in combination with a multiplier transformation technique:
s41: initializing an overall programming vector population Wherein each individual comprises a set of Lagrangian multipliers,/>Is the population scale;
S42: performing individual mutation operation by adopting a formula (10):
s43: performing crossover operation by adopting a formula (11):
s44: obtaining new population after the above process And calculates a corresponding fitness function (9);
S45: mixing the new population and the old population, and reserving the front part with longer total duration of the batching scheme by adopting a greedy optimization mode Individual, forming a new population, and labeling the optimal individual as/>；
S46: calculating population density using equation (8), if the density is less than a preset TF, retainingReinitializing the individuals in the residual population;
s47: if the algorithm does not reach the maximum iteration number, returning to the step S42; otherwise, returning to the optimal individual and the corresponding data coordination value;
s5: and (3) significant error detection:
Computing data coordinated relative residuals And judging the residual error and the Welsch tuning parameter/>If/>Marked as a significant error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011512638.5A CN112734166B (en) | 2020-12-20 | 2020-12-20 | Copper industry data robust coordination and significant error detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011512638.5A CN112734166B (en) | 2020-12-20 | 2020-12-20 | Copper industry data robust coordination and significant error detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112734166A CN112734166A (en) | 2021-04-30 |
| CN112734166B true CN112734166B (en) | 2024-05-03 |
Family
ID=75603934
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011512638.5A Active CN112734166B (en) | 2020-12-20 | 2020-12-20 | Copper industry data robust coordination and significant error detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112734166B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632309A (en) * | 2013-11-05 | 2014-03-12 | 常州大学 | Method for solving unit commitment of wind-fire energy storing-saving and emission-reducing unit |
| CN103824123A (en) * | 2014-01-26 | 2014-05-28 | 河海大学 | Novel distribution network battery energy storage system optimal allocation algorithm |
| CN107392355A (en) * | 2017-06-27 | 2017-11-24 | 北京航空航天大学 | A kind of multimachine field coordination scheduling robust Optimal methods based on differential evolution algorithm |
| CN108519957A (en) * | 2018-02-10 | 2018-09-11 | 大连智慧海洋软件有限公司 | A kind of data coordinating method based on acceleration broad sense reduced gradient |
| CN109686412A (en) * | 2018-12-25 | 2019-04-26 | 阳谷祥光铜业有限公司 | A kind of data harmonization processing method and processing device for metal balance |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10565329B2 (en) * | 2014-06-30 | 2020-02-18 | Evolving Machine Intelligence Pty Ltd | System and method for modelling system behaviour |
-
2020
- 2020-12-20 CN CN202011512638.5A patent/CN112734166B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632309A (en) * | 2013-11-05 | 2014-03-12 | 常州大学 | Method for solving unit commitment of wind-fire energy storing-saving and emission-reducing unit |
| CN103824123A (en) * | 2014-01-26 | 2014-05-28 | 河海大学 | Novel distribution network battery energy storage system optimal allocation algorithm |
| CN107392355A (en) * | 2017-06-27 | 2017-11-24 | 北京航空航天大学 | A kind of multimachine field coordination scheduling robust Optimal methods based on differential evolution algorithm |
| CN108519957A (en) * | 2018-02-10 | 2018-09-11 | 大连智慧海洋软件有限公司 | A kind of data coordinating method based on acceleration broad sense reduced gradient |
| CN109686412A (en) * | 2018-12-25 | 2019-04-26 | 阳谷祥光铜业有限公司 | A kind of data harmonization processing method and processing device for metal balance |
Non-Patent Citations (3)
| Title |
|---|
| 一种热工过程数据协调与显著误差检测同步处理方法;周卫庆;乔宗良;周建新;司风琪;徐治皋;;中国电机工程学报(35);全文 * |
| 基于支持向量机的X射线荧光光谱重金属检测模型的建立;李芳;陆安祥;王纪华;;分析仪器(04);全文 * |
| 铜冶炼过程数据校正方法研究及应用;宋逍翰;《中国优秀硕士学位论文全文数据库工程科技辑》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112734166A (en) | 2021-04-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ye et al. | A novel forecasting method based on multi-order fuzzy time series and technical analysis | |
| CN103258069B (en) | A kind of Forecasting Methodology of steel industry electricity needs | |
| CN103106535B (en) | Method for solving collaborative filtering recommendation data sparsity based on neural network | |
| CN102831269A (en) | Method for determining technological parameters in flow industrial process | |
| CN109647899B (en) | Method for forecasting power consumption of multi-specification rolled pieces in hot rolling and finish rolling process of strip steel | |
| CN106600037B (en) | Multi-parameter auxiliary load prediction method based on principal component analysis | |
| CN107437135B (en) | Novel energy storage type selection method | |
| CN106097055A (en) | Enterprise order processing method under personalized customization demand | |
| CN115423333A (en) | Multi-objective optimization method and system for flow production process based on digital twin | |
| CN108537379A (en) | Adaptive variable weight combination load forecasting method and device | |
| CN104881718A (en) | Regional power business index constructing method based on multi-scale leading economic indicators | |
| CN112016755A (en) | Construction method of universal design cost standardization technology module of power transmission and transformation project construction drawing | |
| CN112884590A (en) | Power grid enterprise financing decision method based on machine learning algorithm | |
| CN116148753A (en) | Intelligent electric energy meter operation error monitoring system | |
| Mathew et al. | Demand forecasting for economic order quantity in inventory management | |
| CN113868938A (en) | Short-term load probability density prediction method, device and system based on quantile regression | |
| CN116468536A (en) | Automatic risk control rule generation method | |
| CN112734166B (en) | Copper industry data robust coordination and significant error detection method | |
| CN105976057B (en) | Medium-and-long-term load prediction method based on improved grey theory | |
| Hallman | A comparative study on Linear Regression and Neural Networks for estimating order quantities of powder blends | |
| CN110196797A (en) | Automatic optimization method and system suitable for credit scoring card system | |
| CN112001551B (en) | Ground and commercial power grid sales electricity quantity prediction method based on large-user electricity quantity information | |
| CN114875196A (en) | Method and system for determining converter tapping quantity | |
| Wang et al. | A probabilistic wind power forecasting approach based on gaussian process regression | |
| CN114331523A (en) | Intellectual property value evaluation system and value evaluation method |
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 |