CN117724431A - Control method and control system for granularity of reaction materials in reaction kettle - Google Patents
Control method and control system for granularity of reaction materials in reaction kettle Download PDFInfo
- Publication number
- CN117724431A CN117724431A CN202410172484.1A CN202410172484A CN117724431A CN 117724431 A CN117724431 A CN 117724431A CN 202410172484 A CN202410172484 A CN 202410172484A CN 117724431 A CN117724431 A CN 117724431A
- Authority
- CN
- China
- Prior art keywords
- granularity
- reaction
- value
- particle size
- reaction kettle
- 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
- 238000006243 chemical reaction Methods 0.000 title claims abstract description 104
- 238000000034 method Methods 0.000 title claims abstract description 47
- 239000000463 material Substances 0.000 title claims abstract description 36
- 239000002245 particle Substances 0.000 claims abstract description 55
- 238000013528 artificial neural network Methods 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 22
- 238000013461 design Methods 0.000 claims abstract description 18
- 239000002243 precursor Substances 0.000 claims abstract description 16
- QGZKDVFQNNGYKY-UHFFFAOYSA-N Ammonia Chemical compound N QGZKDVFQNNGYKY-UHFFFAOYSA-N 0.000 claims description 20
- QAOWNCQODCNURD-UHFFFAOYSA-L Sulfate Chemical compound [O-]S([O-])(=O)=O QAOWNCQODCNURD-UHFFFAOYSA-L 0.000 claims description 20
- 239000007787 solid Substances 0.000 claims description 20
- 239000003513 alkali Substances 0.000 claims description 19
- 239000007788 liquid Substances 0.000 claims description 18
- 239000002019 doping agent Substances 0.000 claims description 11
- VHUUQVKOLVNVRT-UHFFFAOYSA-N Ammonium hydroxide Chemical compound [NH4+].[OH-] VHUUQVKOLVNVRT-UHFFFAOYSA-N 0.000 claims description 10
- 229910021529 ammonia Inorganic materials 0.000 claims description 10
- 235000011114 ammonium hydroxide Nutrition 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000002474 experimental method Methods 0.000 claims description 8
- 230000006870 function Effects 0.000 claims description 7
- 238000003756 stirring Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000010219 correlation analysis Methods 0.000 claims description 6
- 239000007774 positive electrode material Substances 0.000 claims description 5
- 239000000376 reactant Substances 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000005284 excitation Effects 0.000 claims description 4
- 210000002569 neuron Anatomy 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 claims description 3
- 229910001416 lithium ion Inorganic materials 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 6
- 238000009776 industrial production Methods 0.000 abstract description 2
- 230000001276 controlling effect Effects 0.000 description 7
- 230000001105 regulatory effect Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 239000002994 raw material Substances 0.000 description 4
- 230000010354 integration Effects 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- NLXLAEXVIDQMFP-UHFFFAOYSA-N Ammonium chloride Substances [NH4+].[Cl-] NLXLAEXVIDQMFP-UHFFFAOYSA-N 0.000 description 1
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 1
- KFDQGLPGKXUTMZ-UHFFFAOYSA-N [Mn].[Co].[Ni] Chemical compound [Mn].[Co].[Ni] KFDQGLPGKXUTMZ-UHFFFAOYSA-N 0.000 description 1
- 239000010405 anode material Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000000975 co-precipitation Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- QHGJSLXSVXVKHZ-UHFFFAOYSA-N dilithium;dioxido(dioxo)manganese Chemical compound [Li+].[Li+].[O-][Mn]([O-])(=O)=O QHGJSLXSVXVKHZ-UHFFFAOYSA-N 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 229910052744 lithium Inorganic materials 0.000 description 1
- GELKBWJHTRAYNV-UHFFFAOYSA-K lithium iron phosphate Chemical compound [Li+].[Fe+2].[O-]P([O-])([O-])=O GELKBWJHTRAYNV-UHFFFAOYSA-K 0.000 description 1
- 229920000642 polymer Polymers 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004886 process control Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02E60/10—Energy storage using batteries
Landscapes
- Developing Agents For Electrophotography (AREA)
Abstract
The invention discloses a control method of the granularity of a reaction material in a reaction kettle, which comprises the steps of firstly collecting the numerical value of the granularity influence factor of the reaction material in the reaction kettle, carrying out data processing, and then obtaining the weight of the granularity influence factor on the reaction material; acquiring a key granularity influence factor according to the weight and a preset weight threshold; according to the obtained key granularity influencing factors, a BP neural network prediction model is established, and the granularity of the reaction materials is predicted according to the model to obtain a granularity prediction value; and comparing the particle size predicted value with a particle size target design value of the reaction kettle, and controlling related parameters of the reaction kettle according to a comparison result to realize particle size control. The invention not only can improve the applicability, the intelligent operation and the automation level, but also has important significance for improving the product quality of the industrial production precursor and reducing the production cost.
Description
Technical Field
The invention belongs to the technical field of precision control of reaction kettles, and particularly relates to a method and a system for controlling granularity of materials in a reaction kettle.
Background
Along with the rising of the sales of electric vehicles in China, the whole industrial chain at the upstream and downstream is rapidly developed. As the most expensive and major component of electric vehicles, the cost of the positive electrode material accounts for more than 40% of the total cost of the battery. That is, the positive electrode material is one of important elements for the development of the left and right electric automobile industries. Among the elements constituting the positive electrode material, lithium manganate, lithium cobaltate, lithium iron phosphate and ternary materials (polymers of nickel cobalt manganese) are most common. The precursor is critical to the production of ternary materials, as the quality of the precursor (morphology, particle size distribution, specific surface area, impurity content, tap density, etc.) directly determines the physicochemical index of the final sintered product. In the process of precursor industrialization, the synthesis of the precursor is realized by a reaction kettle through coprecipitation reaction of a sulfate solution, an alkali solution and an ammonia solution according to a certain proportion by using a metering pump. As the core equipment of the reaction kettle, various factors such as solution flow, temperature, pH and stirring rate of the reaction kettle can influence the granularity of the precursor. Because the granularity change in the reaction kettle is affected by a plurality of factors, the granularity change cannot be determined by a single characteristic value. If an accurate mathematical modeling is performed manually for one reaction in one factory, not only is the modeling difficult and the workload great, but also the modeling cannot be applied to another factory, and the universality is poor.
Disclosure of Invention
The invention aims to solve the technical problems, overcome the defects and the defects in the background technology, improve the universality of the method and aim at different factory production, and therefore, the invention provides a control method and a control system for the granularity of reaction materials in a reaction kettle.
In order to solve the technical problems, the technical scheme provided by the invention is a method for controlling the granularity of reaction materials in a reaction kettle, which comprises the following steps:
step 1: collecting the value of the granularity influencing factor of the reaction materials of the reaction kettle, and carrying out data processing on the value of the granularity influencing factor;
step 2: acquiring the weight of the influence of the particle size influence factors on the granularity of the reaction materials according to the acquired numerical value of each granularity influence factor;
step 3: selecting granularity influence factors according to the weight obtained in the step 2 and a preset weight threshold value, and obtaining key granularity influence factors;
step 4: according to the obtained key granularity influencing factors, a BP neural network prediction model is established, and the granularity of the reaction materials is predicted according to the BP neural network prediction model, so that a granularity predicted value is obtained;
step 5: and comparing the particle size predicted value with a particle size target design value of the reaction kettle, and controlling related parameters of the reaction kettle according to a comparison result to realize particle size control.
In the above control method, preferably, in step 1, the particle size influencing factor includes at least two of a feed liquid flow rate of the raw material, a feed liquid concentration of the raw material, a reaction pH value, a reaction temperature, a reaction stirring rate, doping (doping agent), and a solid content. The raw material feed liquid is one or more of sulfate solution, alkali liquor and ammonia water. Several granularity influencing factors can be used for data preprocessing, data cleaning and data integration preferably due to factors such as inconsistent data, incomplete data, information errors and the like.
In the above control method, preferably, in step 1, the reaction material refers to a ternary precursor of a positive electrode material of a lithium ion battery, and the particle size influencing factors include sulfate flow, alkali solution flow, ammonia water flow, reaction temperature, reaction pH, reaction stirring rate, doping and solid content. More preferably including sulfate flow, lye flow, ammonia flow, doping and solids content.
The influence of each granularity influence factor on granularity is different, so that the method has more universality, achieves the effect of intelligent selection, and also prevents too many granularity influence factors of an input prediction algorithm from causing too slow establishment time of an algorithm model, and is preferable, and the granularity influence weight of each granularity influence factor on the granularity of a reaction material is carried out by using a gray correlation analysis algorithm. Because the factors influencing the granularity change in the reaction kettle are complicated, the weights of the factors influencing the granularity change in the reaction kettle are different, and the relation between the granularity influencing factors and the predicted objects can be automatically, accurately and effectively extracted according to the granularity influencing factor data by using a gray correlation algorithm, and the requirement on the data is low.
In the above control method, preferably, the gray correlation analysis algorithm includes data initialization and gray correlation coefficient calculation steps:
the data initialization refers to normalization processing of the granularity influence factors;
the gray correlation coefficient is calculated according to the following formula:
；
wherein X is i (k) For the column to be analyzed, xo (k) is a reference column, ρ is a resolution factor, and the range of ρ is 0 to 1.
In the above control method, preferably, in step 3, the weight threshold is determined by:
setting a threshold value increasing experiment, namely setting the threshold value to be continuously increased, selecting a granularity influencing factor of which the gray correlation coefficient exceeds the threshold value as input of a neural network model when the threshold value is set once, taking granularity as output of the neural network model, training, and obtaining average relative errors of threshold value prediction set each time after multiple training experiments are carried out; and finally, selecting a corresponding threshold value with the smallest predicted average relative error as a set weight threshold value.
In the above control method, preferably, in step 3, the key granularity influencing factor is a granularity influencing factor selected from gray correlation coefficients greater than a weight threshold.
In the above control method, preferably, the establishing of the BP neural network prediction model includes: and selecting the optimal number of neurons of the hidden layer, and continuously adjusting weights among layers of the neural network by adopting hyperbolic tangent Sigmoid functions as excitation functions of the hidden layer and the output layer, so that errors between the output of the neural network and actual measurement results are minimized. By heuristics, the most favorable number of neurons in the hidden layer is generally between 1-3 times of the variable number of the input layer, and experiments can be gradually increased from less to more in the range of 10-30, until the model effect is no longer improved.
In the above control method, preferably, the controlling the relevant parameters of the reaction kettle according to the comparison result specifically includes the following operations: if the predicted value of the particle size predicted value is larger than the target design value of the particle size of the reaction kettle, at least one measure of increasing sulfate flow, ammonia water flow, reducing alkali liquid flow, reducing the dosage of doping agent, reducing pH value and reducing solid content is adopted; if the predicted value of the granularity predicted value is smaller than the target design value of the granularity of the reaction kettle, at least one measure of reducing the sulfate flow, the ammonia water flow, increasing the alkali liquid flow, increasing the dosage of the doping agent, increasing the pH value and improving the solid content is adopted.
As a general technical concept, the present invention also provides a control system of granularity of a reaction material in a reaction kettle, including:
the data acquisition module is used for acquiring the value of the granularity influencing factor of the reaction materials of the reaction kettle and carrying out data processing on the value of the granularity influencing factor;
the data processing module is used for acquiring the weight of each particle size influence factor on the particle size influence of the reactant according to the acquired numerical value of each particle size influence factor, selecting the particle size influence factor according to the acquired weight and a preset weight threshold value, and acquiring and outputting a key particle size influence factor;
the data processing module further comprises a BP neural network prediction model established according to the input key granularity influence factors, wherein the BP neural network prediction model is used for predicting granularity of the reaction materials and outputting granularity prediction values;
the data processing module further comprises a step of outputting different control parameters according to a comparison result of the granularity predicted value and the granularity target design value of the reaction kettle;
and the control module is used for adjusting and controlling the process parameter adjusting device of the reaction kettle according to the control parameters output by the data processing module.
Compared with the prior art, the invention has the beneficial effects that: the invention analyzes the demand of industrial automatic production, and designs a control system and a control method for realizing the granularity of the reaction materials in the reaction kettle according to the result of demand analysis and the granularity prediction intelligent algorithm designed in the invention. The method can improve universality, operation intelligence and automation level, and has important significance for improving the product quality of industrial production precursors and reducing the production cost.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the material particle size prediction of the reaction kettle in the embodiment of the invention.
FIG. 2 is a graph showing the relationship between threshold selection and average relative error in an embodiment of the present invention.
FIG. 3 is a schematic diagram of a BP neural network prediction model according to an embodiment of the present invention; wherein x is input, O is output, phi and phi are hidden layer and output layer excitation functions respectively, w is weight, and theta and alpha are hidden layer and output layer thresholds respectively.
Detailed Description
The present invention will be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments are shown, for the purpose of illustrating the invention, but the scope of the invention is not limited to the specific embodiments shown.
Unless defined otherwise, all technical and scientific terms used hereinafter have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the scope of the present invention.
Unless otherwise specifically indicated, the various raw materials, reagents, instruments, equipment and the like used in the present invention are commercially available or may be prepared by existing methods.
Examples:
the invention discloses a method for predicting and controlling the dynamic particle size of ternary precursor particles of a lithium ion battery anode material in real time, which is shown in figure 1, and specifically comprises the following steps:
step 1:
the method comprises the steps of collecting data of particle size influence factors of reaction materials at the current moment of a reaction kettle and processing the data, and specifically comprises the following steps:
step 1.1:
data processing generally includes data preprocessing, data cleansing, data integration, and the like. Because of factors such as inconsistent data, incomplete data, information errors and the like of the influence factors of the reaction kettles, the collected data needs to be subjected to data preprocessing, data cleaning and data integration.
The particle size influencing factors selected in the embodiment comprise sulfate flow, alkali liquid flow, ammonia water flow, reaction temperature, reaction pH value, reaction stirring rate, doping and solid content.
Step 2:
acquiring the weight of the influence of the particle size influence factors on the granularity of the reaction materials according to the acquired numerical value of each granularity influence factor; in the embodiment, a gray correlation analysis method is used for calculating the correlation coefficient between each granularity influence factor and granularity, and the specific steps are as follows:
step 2.1:
data initialization: carrying out normalization (normalization) treatment on data of sulfate flow, alkali liquid flow, ammonia water flow, reaction temperature, reaction pH value, reaction stirring rate, doping and solid content, reducing the difference of absolute values of the data, and unifying the absolute values to be in an approximate range;
step 2.2:
gray correlation coefficient calculation: calculating gray correlation coefficients of each particle size influence factor and the particle size of the reaction materials according to the following formula;
；
wherein X is i (k) For the column to be analyzed, xo (k) is a reference column, ρ is a resolution factor, and ρ ranges from 0 to 1, preferably 0.5 in this embodiment. The gray correlation coefficients of each particle size influencing factor calculated in this example and the reactant material particle size D10 are shown in table 1 below:
table 1: gray correlation value of granularity D10 and each granularity influence factor
Step 3:
screening gray correlation coefficients above the weight threshold according to the weight obtained in the step 2 and a preset weight threshold, selecting granularity influence factors, and obtaining key granularity influence factors (namely main factors);
step 3.1:
when the weight threshold is set to 0.5, 6 factors of sulfate flow, alkali liquid flow, ammonia flow, doping, reaction pH value and solid content are selected as granularity influencing factors, data corresponding to the 6 granularity influencing factors are used as input of a neural network model, granularity is used as output for training, and a predicted average relative error delta is obtained after multiple tests are carried out 1 。
Step 3.2:
when the weight threshold is set to 0.6, 5 factors of sulfate flow, alkali liquid flow, ammonia flow, doping and solid content are selected as granularity influencing factors, data corresponding to the 5 granularity influencing factors are used as input of a neural network model, granularity is used as output for training, and the predicted average relative error delta is obtained after multiple experiments are carried out 2 。
Step 3.3:
according to the set threshold increment experiment, the prediction accuracy of the neural network can be remarkably improved along with the increase of the threshold, namely, the elimination of the secondary factors, but the higher the threshold is, the better the prediction accuracy of the neural network is, the lower the prediction accuracy of the neural network is possibly. And by analogy, selecting a weight threshold value with the smallest average relative error of the predictions as a set optimal weight threshold value, and taking a granularity influence factor larger than the optimal weight threshold value as a key granularity influence factor. In this embodiment, the optimal weight threshold value is 0.6 to achieve the best effect. At this time, the sulfate flow, alkali liquid flow, ammonia flow, doping and solid content are selected as key granularity influencing factors.
Step 4:
and taking the key granularity influence factors as input and granularity as output, and establishing a BP neural network prediction model.
And selecting the optimal number of neurons of the hidden layer, and continuously adjusting weights among layers of the neural network by adopting hyperbolic tangent Sigmoid functions as excitation functions of the hidden layer and the output layer, so that errors between the output and actual results of the neural network are minimized. The invention sets the allowable precision eps of the neural network to be 10 -5 And the maximum learning times are 900, and training samples are 500 groups.
Step 5:
and predicting granularity by using the established BP neural network prediction model. Predicting by using a trained BP neural network prediction model, using data of key granularity influence factors as input for the learned model, verifying the correctness of the BP neural network prediction model, and outputting a granularity prediction value.
By comparing the actual data with the prediction experiment, we found that the prediction accuracy was 89.95% by using only the conventional BP neural network model (see Table 2 below).
Table 2: partial prediction data of conventional BP neural network is compared with original data
The comparison of the predicted data of the prediction model and the original data by adopting the method provided by the embodiment of the invention is shown in the following table 3, and the prediction precision can reach 95.52%.
Table 3: the method based on BP neural network of the invention compares partial predicted data with original data
From the above, the innovation of the method is mainly that aiming at the situation that no method or model for predicting the granularity of the precursor exists at present, in order to meet the urgent need of precursor production and process control on granularity data, the invention provides a method for predicting and regulating the granularity of the precursor, which comprises the following steps: the method combines BP neural network and gray correlation analysis to design an efficient granularity prediction model, has higher prediction precision and intelligent level, and provides a new idea for improving the accuracy of granularity prediction.
The conventional precursor granularity regulation is performed under the guidance of theoretical analysis and practical experience, the granularity can be accurately determined by a calculation formula without fixation, and only the flow, the temperature and the like are evaluated and regulated by the granularity collected manually, so that the regulation measures are quite rough. In addition, the conventional method is difficult to comprehensively consider complex nonlinear relations and interactions among influencing factors, so that the deviation from the actually required granularity is large, and the regulation and control effects are general. The control method is applied to the synthesis and precipitation process of the ternary precursor, and if the predicted value of the granularity is larger than the target design value of the granularity of the reaction kettle, at least one measure of increasing sulfate flow, ammonia water flow, reducing alkali liquid flow, reducing the dosage of doping agent, reducing pH value and reducing the solid content is adopted; if the predicted value of the granularity predicted value is smaller than the target design value of the granularity of the reaction kettle, at least one of the measures of reducing the sulfate flow, the ammonia flow, increasing the alkali liquid flow, increasing the dosage of the doping agent, increasing the pH value and improving the solid content requirement is adopted, so that the granularity prediction precision in the preparation process of the ternary precursor can be improved as much as possible, and the method is particularly suitable for the granularity control requirement prediction of the reaction kettle, which is influenced by various uncertainty factors and is difficult to establish a determined prediction model.
Taking some experimental data in table 3 as an example, when the particle size target design value D10 of the target product is 9 micrometers, the measured variable parameters are shown in the following table 4, at this time, the D10 predicted by the neural network model according to the method of the invention is 8.61 micrometers, that is, the predicted particle size value is smaller than the particle size target design value of the reaction kettle, at least one of the measures of reducing sulfate flow, ammonia flow, increasing alkali liquor flow, increasing doping agent dosage, increasing pH value and increasing solid content is needed, in this embodiment, the BP neural network self-regulating measure is adopted, the sulfate flow, ammonia flow, increasing alkali liquor flow, increasing doping agent dosage, increasing pH value and improving solid content are synchronously reduced, the parameters after regulation are controlled as shown in the following table 5, and the final measured particle size is 8.95 micrometers.
Table 4: parameter value and granularity predicted value before regulation under certain application scene
Table 5: parameter value and granularity actual measurement value regulated and controlled under certain application scene
Taking part of experimental data in table 3 as an example, when the particle size target design value D10 of the target product is 11 micrometers, the measured variable parameters are shown in the following table 6, and at this time, D10 predicted by the neural network model according to the method of the invention is 11.59 micrometers, that is, the predicted value of the particle size predicted value is greater than the particle size target design value of the reaction kettle, at least one measure of increasing sulfate flow, ammonia flow, reducing alkali flow, reducing the dosage of doping agent, reducing pH value and reducing solid content is required; in this example, the BP neural network adopts self-regulating measures, which synchronously increase sulfate flow, ammonia flow, alkali liquor flow, doping agent consumption, pH value and solid content, and the regulated parameters are controlled as shown in the following table 7, and the final measured granularity is 11.15 micrometers.
Table 6: parameter value and granularity predicted value before regulation under certain application scene
Table 7: parameter value and granularity predicted value before regulation under certain application scene
The embodiment also provides a control system for the granularity of the reaction materials in the reaction kettle, which adopts the method, and the control system comprises:
the data acquisition module is used for acquiring the value of the granularity influencing factor of the reaction materials of the reaction kettle and carrying out data processing on the value of the granularity influencing factor;
the data processing module is used for acquiring the weight of each particle size influence factor on the particle size influence of the reactant according to the acquired numerical value of each particle size influence factor, selecting the particle size influence factor according to the acquired weight and a preset weight threshold value, and acquiring and outputting a key particle size influence factor;
the data processing module further comprises a BP neural network prediction model established according to the input key granularity influence factors, wherein the BP neural network prediction model is used for predicting granularity of the reaction materials and outputting granularity prediction values;
the data processing module further comprises a step of outputting different control parameters according to a comparison result of the granularity predicted value and the granularity target design value of the reaction kettle;
and the control module is used for adjusting and controlling the process parameter adjusting device of the reaction kettle according to the control parameters output by the data processing module.
According to the control method and the control system, data are collected in real time through historical data collection, data are analyzed through an algorithm, correlation coefficients among various data are calculated, a threshold value is set according to the prediction relative error, influence factors with the correlation coefficients being larger than an optimal threshold value are selected as main factors, a BP algorithm prediction model is built according to the main factors, and then granularity is predicted by the data and the built algorithm model and regulated and controlled based on the prediction values. The invention improves the control precision of the granularity prediction of the reaction kettle, and ensures that the granularity prediction of the reaction kettle has better control performance.
Claims (10)
1. The control method of the granularity of the reaction materials in the reaction kettle is characterized by comprising the following steps:
step 1: collecting the value of the granularity influencing factor of the reaction materials of the reaction kettle, and carrying out data processing on the value of the granularity influencing factor;
step 2: acquiring the weight of the influence of the particle size influence factors on the granularity of the reaction materials according to the acquired numerical value of each granularity influence factor;
step 3: selecting granularity influence factors according to the weight obtained in the step 2 and a preset weight threshold value, and obtaining key granularity influence factors;
step 4: according to the obtained key granularity influencing factors, a BP neural network prediction model is established, and the granularity of the reaction materials is predicted according to the BP neural network prediction model, so that a granularity predicted value is obtained;
step 5: and comparing the particle size predicted value with a particle size target design value of the reaction kettle, and controlling related parameters of the reaction kettle according to a comparison result to realize particle size control.
2. The control method according to claim 1, wherein in step 1, the particle size influencing factor includes at least two of a feed liquid flow rate of a feed liquid, a feed liquid concentration of a feed liquid, a reaction pH value, a reaction temperature, a reaction stirring rate, doping, and a solid content.
3. The control method according to claim 2, wherein in step 1, the reaction material is a ternary precursor of a positive electrode material of a lithium ion battery, and the particle size influencing factors include sulfate flow, alkali solution flow, ammonia flow, reaction temperature, reaction pH, reaction stirring rate, doping and solid content.
4. The control method of claim 1, wherein the weight of each particle size impact factor on reactant material particle size impact is performed using a gray correlation analysis algorithm.
5. The control method according to claim 4, wherein the gray correlation analysis algorithm includes a data initialization and gray correlation coefficient calculation step:
the data initialization refers to normalization processing of the granularity influence factors;
the gray correlation coefficient is calculated according to the following formula:
；
wherein Xi (k) is a column to be analyzed, xo (k) is a reference column, ρ is a resolution coefficient, and ρ ranges from 0 to 1.
6. The control method according to claim 5, characterized in that in step 3, the weight threshold is determined by:
setting a threshold value increasing experiment, namely setting the threshold value to be continuously increased, selecting a granularity influencing factor of which the gray correlation coefficient exceeds the threshold value as input of a neural network model when the threshold value is set once, taking granularity as output of the neural network model, training, and obtaining average relative errors of threshold value prediction set each time after multiple training experiments are carried out; and finally, selecting a corresponding threshold value with the smallest predicted average relative error as a set weight threshold value.
7. The control method according to claim 6, wherein in the step 3, the key granularity influencing factor is a granularity influencing factor with a gray correlation coefficient larger than a weight threshold value.
8. The control method according to any one of claims 1 to 7, characterized in that the establishment of the BP neural network prediction model includes: and selecting the optimal number of neurons of the hidden layer, and continuously adjusting weights among layers of the neural network by adopting hyperbolic tangent Sigmoid functions as excitation functions of the hidden layer and the output layer, so that errors between the output of the neural network and actual measurement results are minimized.
9. The control method according to any one of claims 1 to 7, wherein the controlling the relevant parameters of the reaction vessel according to the comparison result specifically comprises the following operations: if the predicted value of the particle size predicted value is larger than the target design value of the particle size of the reaction kettle, at least one measure of increasing sulfate flow, ammonia water flow, reducing alkali liquid flow, reducing the dosage of doping agent, reducing pH value and reducing solid content is adopted; if the predicted value of the granularity predicted value is smaller than the target design value of the granularity of the reaction kettle, at least one measure of reducing the sulfate flow, the ammonia water flow, increasing the alkali liquid flow, increasing the dosage of the doping agent, increasing the pH value and improving the solid content is adopted.
10. A control system for the particle size of a reaction mass in a reaction vessel, comprising:
the data acquisition module is used for acquiring the value of the granularity influencing factor of the reaction materials of the reaction kettle and carrying out data processing on the value of the granularity influencing factor;
the data processing module is used for acquiring the weight of each particle size influence factor on the particle size influence of the reactant according to the acquired numerical value of each particle size influence factor, selecting the particle size influence factor according to the acquired weight and a preset weight threshold value, and acquiring and outputting a key particle size influence factor;
the data processing module further comprises a BP neural network prediction model established according to the input key granularity influence factors, wherein the BP neural network prediction model is used for predicting granularity of the reaction materials and outputting granularity prediction values;
the data processing module further comprises a step of outputting different control parameters according to a comparison result of the granularity predicted value and the granularity target design value of the reaction kettle;
and the control module is used for adjusting and controlling the process parameter adjusting device of the reaction kettle according to the control parameters output by the data processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410172484.1A CN117724431A (en) | 2024-02-07 | 2024-02-07 | Control method and control system for granularity of reaction materials in reaction kettle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410172484.1A CN117724431A (en) | 2024-02-07 | 2024-02-07 | Control method and control system for granularity of reaction materials in reaction kettle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117724431A true CN117724431A (en) | 2024-03-19 |
Family
ID=90207288
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410172484.1A Pending CN117724431A (en) | 2024-02-07 | 2024-02-07 | Control method and control system for granularity of reaction materials in reaction kettle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117724431A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412489A (en) * | 2013-08-12 | 2013-11-27 | 东北大学 | Ore grinding granularity online prediction system and method |
| CN103793887A (en) * | 2014-02-17 | 2014-05-14 | 华北电力大学 | Short-term electrical load on-line predicting method based on self-adaptation enhancing algorithm |
| CN107767022A (en) * | 2017-09-12 | 2018-03-06 | 重庆邮电大学 | A kind of Dynamic Job-shop Scheduling rule intelligent selecting method of creation data driving |
| CN110070104A (en) * | 2019-03-22 | 2019-07-30 | 腾讯科技（上海）有限公司 | A kind of hyper parameter determines method, apparatus and server |
| CN112287598A (en) * | 2020-09-28 | 2021-01-29 | 山西漳山发电有限责任公司 | Fly ash carbon content prediction method based on particle swarm parameter optimization |
-
2024
- 2024-02-07 CN CN202410172484.1A patent/CN117724431A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412489A (en) * | 2013-08-12 | 2013-11-27 | 东北大学 | Ore grinding granularity online prediction system and method |
| CN103793887A (en) * | 2014-02-17 | 2014-05-14 | 华北电力大学 | Short-term electrical load on-line predicting method based on self-adaptation enhancing algorithm |
| CN107767022A (en) * | 2017-09-12 | 2018-03-06 | 重庆邮电大学 | A kind of Dynamic Job-shop Scheduling rule intelligent selecting method of creation data driving |
| CN110070104A (en) * | 2019-03-22 | 2019-07-30 | 腾讯科技（上海）有限公司 | A kind of hyper parameter determines method, apparatus and server |
| CN112287598A (en) * | 2020-09-28 | 2021-01-29 | 山西漳山发电有限责任公司 | Fly ash carbon content prediction method based on particle swarm parameter optimization |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110320806B (en) | Sewage treatment process self-adaptive prediction control method based on integrated instant learning | |
| AU2021101438A4 (en) | Adaptive control method and system for aeration process | |
| CN110320335B (en) | Multi-element robust soft measurement method for effluent quality index of sewage treatment | |
| CN111533290B (en) | Method for generating optimal operation plan library of sewage treatment process and applying complex scene | |
| CN101863088A (en) | Method for forecasting Mooney viscosity in rubber mixing process | |
| CN110981021B (en) | Intelligent dosing system and method for advanced wastewater oxidation treatment based on fuzzy BP neural network | |
| CN115793471B (en) | Adjustable control method and system based on sewage treatment monitoring | |
| CN110357236B (en) | Sewage plant intelligent control method based on mutation inversion effluent prediction model | |
| CN110390132B (en) | Nonferrous metallurgy unit procedure digitalization and modeling method based on process state space | |
| CN114781249A (en) | High-density clarification tank dosage prediction and control method based on multidimensional scoring model | |
| CN104263960B (en) | On-line prediction method for replacement rate and gold mud grade in gold hydrometallurgy replacement process | |
| CN117724431A (en) | Control method and control system for granularity of reaction materials in reaction kettle | |
| CN115202190B (en) | Working condition dividing and identifying method for adjusting potential based on process index | |
| CN114871000B (en) | Flotation dosing self-adaptive regulation and control method | |
| CN112899496B (en) | Method and system for estimating concentration of copper ions at inlet of zinc hydrometallurgy purification cobalt removal process | |
| CN109231389A (en) | A kind of multi-state coordination optimizing method for heavy metal containing wastewater treatment process | |
| CN115291519A (en) | Intelligent optimization control method for ore grinding process | |
| CN114841277A (en) | Method and system for predicting yield of converter tapping alloy | |
| CN113221436A (en) | Soft measurement method for sewage suspended matter concentration based on improved RBF neural network | |
| 高学金 et al. | An optimal control strategy combining SVM with RGA for improving fermentation titer | |
| CN106127326B (en) | Chemical material processing melt index forecasting method | |
| Ling et al. | A soft sensor based on influent mode discrimination neural network for a wastewater treatment process | |
| Juuso et al. | Intelligent analysers and dynamic simulation in a biological water treatment process | |
| Long et al. | pH prediction of a neutral leaching process using adaptive-network-based fuzzy inference system and reaction kinetics | |
| CN116976148B (en) | Method and system for monitoring ion content change in copper electrolysis process |
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 |