CN-122020940-A - Cold hydrogenation process parameter prediction method, optimization method, device and equipment

CN122020940ACN 122020940 ACN122020940 ACN 122020940ACN-122020940-A

Abstract

The invention discloses a cold hydrogenation process parameter prediction method, an optimization method, a device and equipment, wherein the cold hydrogenation process parameter prediction method comprises the steps of obtaining initial process parameters of a cold hydrogenation process flow; the method comprises the steps of obtaining a pre-established cold hydrogenation process parameter composite model, wherein the cold hydrogenation process parameter composite model comprises a front-stage mechanism model, a reactor big data model and a rear-stage mechanism model, and simulating the whole cold hydrogenation process flow by utilizing the cold hydrogenation process parameter composite model based on the initial process parameters to obtain the predicted parameters of the cold hydrogenation process flow. The method can simulate and predict the whole process flow by acquiring initial process parameters and utilizing a composite model consisting of a front-stage mechanism model, a reactor big data model and a back-stage mechanism model, thereby realizing accurate prediction and analysis of process performance.

Inventors

SONG JIAN
Zhang mengze
FAN XIECHENG
ZHAO HUITING
JIANG LAN
LIU CHONG
LIU YANYU

Assignees

内蒙古新特硅材料有限公司

Dates

Publication Date: 20260512
Application Date: 20241112

Claims (11)

1. A method for predicting a cold hydrogenation process parameter, the method comprising: Obtaining initial technological parameters of a cold hydrogenation process flow, wherein the initial technological parameters are technological conditions and operation parameters which are preset and used for simulating and predicting the performance of the process flow when the cold hydrogenation process flow starts; acquiring a pre-established cold hydrogenation process parameter composite model, wherein the cold hydrogenation process parameter composite model comprises a front-stage mechanism model, a reactor big data model and a rear-stage mechanism model; based on the initial technological parameters, the whole cold hydrogenation technological process is simulated by utilizing a cold hydrogenation technological parameter composite model to obtain predicted parameters of the cold hydrogenation technological process.
2. The method according to claim 1, wherein the method further comprises: collecting historical process parameters and experimental analysis data of a cold hydrogenation process flow; According to the historical process parameters and experimental analysis data, an initial mechanism model of a cold hydrogenation whole process is established based on chemical thermodynamics, reaction mechanism, rectification theory and fluidized bed reaction dynamics, wherein the initial mechanism model at least comprises a heat exchanger module, a fluidized bed reactor module, a quench tower module, a cooler module, a flash tank module, a rectification tower module, a hydrogen compressor module and a pump module; carrying out correlation analysis on the historical process parameters, and screening out key parameters of the reactor, wherein the key parameters of the reactor are parameters affecting the performance of the fluidized bed reactor; According to the key parameters of the reactor in the historical process parameters, training by using a neural network to obtain a big data model of the reactor; Replacing the fluidized bed reactor module in the initial mechanism model with the reactor big data model to obtain the cold hydrogenation process parameter composite model, The front-stage mechanism model comprises the heat exchanger module, and the rear-stage mechanism model comprises the quenching tower module, the cooler module, the flash tank module, the rectifying tower module, the hydrogen compressor module and the pump module.
3. The method according to claim 2, wherein the correlation analysis of the historical process parameters screens out reactor key parameters, specifically comprising: Carrying out pearson correlation coefficient analysis algorithm on the collected historical process parameters to carry out correlation analysis calculation, and screening to obtain the key parameters of the reactor; the expression of the pearson correlation coefficient analysis algorithm is the following expression (1): Wherein r is a correlation result, n, xi, x, S x represents the number of samples of the operation parameter, the ith operation parameter, the average value of the operation parameter, and the standard deviation of the operation parameter, y i is the ith product quality data, y is the average value of the product quality data, and S y is the standard deviation of the product quality data.
4. The method according to claim 2, wherein the expression of the reactor big data model is the following expression (2): [Z]=Φ(X,Y) (2) Wherein Φ is a formalized equation of the big data model, X is an input variable comprising the above, Y is a hyper-parameter of the big data model, and Z comprises the purity of the silane at the outlet of the fluidized bed.
5. The method according to claim 2, wherein said replacing the fluidized bed reactor module in the initial mechanism model with the reactor big data model results in the cold hydrogenation process parameter composite model, specifically comprising: and taking the output parameters of the reactor big data model as the input parameters of the back-end mechanism model.
6. A method for optimizing cold hydrogenation process parameters, the method comprising: S1, acquiring initial technological parameters of all equipment of a polysilicon production system in the production process; S2, obtaining predicted process parameters by using the cold hydrogenation process parameter prediction method according to any one of claims 1 to 5 based on the initial process parameters; S3, under the condition that the predicted technological parameters do not meet the optimization target, adjusting the initial technological parameters, updating the initial technological parameters, returning to execute the step S2, And outputting the initial process parameters and the predicted process parameters until the predicted process parameters meet an optimization target. S4, adjusting actual process parameters of all equipment in the production process of the polysilicon production system according to the initial process parameters.
7. A cold hydrogenation process parameter prediction apparatus, the apparatus comprising: the first acquisition module is used for acquiring initial process parameters of the cold hydrogenation process flow, wherein the initial process parameters are process conditions and operation parameters which are preset when the cold hydrogenation process flow starts and are used for simulating and predicting the performance of the process flow; The second acquisition module is used for acquiring a pre-established cold hydrogenation process parameter composite model, wherein the cold hydrogenation process parameter composite model comprises a front-stage mechanism model, a reactor big data model and a rear-stage mechanism model; the prediction module is respectively connected with the first acquisition module and the second acquisition module and is used for simulating the whole cold hydrogenation process flow by utilizing a cold hydrogenation process parameter composite model based on the initial process parameters to obtain the prediction parameters of the cold hydrogenation process flow.
8. The apparatus of claim 7, wherein the apparatus further comprises: the data acquisition module is used for acquiring historical process parameters and experimental analysis data of the cold hydrogenation process flow; The first creation module is connected with the data acquisition module and is used for establishing an initial mechanism model of a cold hydrogenation whole process based on chemical thermodynamics, reaction mechanism, rectification theory and fluidized bed reaction dynamics according to the historical process parameters and experimental analysis data, wherein the initial mechanism model at least comprises a heat exchanger module, a fluidized bed reactor module, a quench tower module, a cooler module, a flash tank module, a rectification tower module, a hydrogen compressor module and a pump module; The screening module is connected with the data acquisition module and used for carrying out correlation analysis on the historical process parameters and screening out key parameters of the reactor, wherein the key parameters of the reactor are parameters affecting the performance of the fluidized bed reactor; The second creation module is connected with the screening module and is used for obtaining the reactor big data model by training through a neural network according to the key parameters of the reactor in the historical process parameters; the third creation module is respectively connected with the first creation module and the second creation module and is used for replacing the fluidized bed reactor module in the initial mechanism model with the reactor big data model to obtain the cold hydrogenation process parameter composite model, The front-stage mechanism model comprises the heat exchanger module, and the rear-stage mechanism model comprises the quenching tower module, the cooler module, the flash tank module, the rectifying tower module, the hydrogen compressor module and the pump module.
9. The apparatus of claim 8, wherein the screening module comprises: The screening unit is used for carrying out correlation analysis and calculation on the collected historical process parameters by a Pelson correlation coefficient analysis algorithm, and screening to obtain the key parameters of the reactor; the expression of the pearson correlation coefficient analysis algorithm is the following expression (1): Wherein r is a correlation result, n, xi, x, sx represents the number of samples of the operation parameter, the ith operation parameter, the average value of the operation parameter, and the standard deviation of the operation parameter, y i is the ith product quality data, y is the average value of the product quality data, and S y is the standard deviation of the product quality data.
10. A cold hydrogenation process parameter optimization apparatus, the apparatus comprising: The acquisition module is used for acquiring initial technological parameters of all equipment in the production process of the polysilicon production system; The cold hydrogenation process parameter predicting device according to any one of claims 6 to 8, connected to the acquisition module, and configured to obtain a predicted process parameter based on the set initial process parameter; The control module is respectively connected with the setting module and the cold hydrogenation process parameter prediction device and is used for adjusting the initial process parameter and updating the initial process parameter under the condition that the predicted process parameter does not meet the optimization target, Controlling the cold hydrogenation process parameter prediction device to obtain a predicted process parameter based on the initial process parameter, Outputting the initial process parameters until the predicted process parameters meet an optimization target; And the optimizing module is connected with the control module and is used for adjusting the actual technological parameters of all equipment in the production process of the polysilicon production system according to the initial technological parameters.
11. An electronic device, characterized in that, the apparatus includes the apparatus comprising: a processor and a memory storing computer program instructions; The processor, when executing the computer program instructions, implements the method of any of claims 1-6.

Description

Cold hydrogenation process parameter prediction method, optimization method, device and equipment Technical Field The invention belongs to the technical field of cold hydrogenation, and particularly relates to a cold hydrogenation process parameter prediction method, an optimization method, a device and equipment. Background The hydrogenation technology is used as a technology for effectively treating polycrystalline silicon byproducts, the main stream technology is a cold hydrogenation technology, silicon powder, a catalyst, silicon tetrachloride and hydrogen are introduced into a fluidized bed reactor in the cold hydrogenation production process, the silicon powder, the catalyst, the silicon tetrachloride and the hydrogen are converted into trichlorosilane at a certain temperature and under a certain pressure, and then the trichlorosilane is subjected to multistage rectification and purification. In the production process, technological parameters are required to be continuously optimized to realize higher conversion rate under lower energy consumption. Silicon chloride (SiHCl 3) is an important intermediate in the production of polysilicon, and its conversion rate has a key impact on production efficiency and product quality. However, due to the limitation of the prior art means, the trichlorosilane conversion rate cannot be measured in real time, so that the adjustment of the process parameters lacks timeliness and the rapid response to the production change cannot be realized. Meanwhile, when the yield of components is predicted, the method based on pure mechanism modeling or pure big data modeling has the problems of poor convergence and low prediction accuracy, influences the reliability of a prediction result, and cannot effectively guide the optimization and adjustment of technological parameters. Therefore, the prior art cannot accurately predict the process parameters of the cold hydrogenation process flow. Disclosure of Invention The technical problem to be solved by the invention is to provide a cold hydrogenation process parameter prediction method, an optimization method, a device and equipment aiming at the defects in the prior art, and the method can improve the accuracy of cold hydrogenation process parameter prediction. In a first aspect, an embodiment of the present invention provides a method for predicting a cold hydrogenation process parameter, the method comprising: Obtaining initial technological parameters of a cold hydrogenation process flow, wherein the initial technological parameters are technological conditions and operation parameters which are preset and used for simulating and predicting the performance of the process flow when the cold hydrogenation process flow starts; acquiring a pre-established cold hydrogenation process parameter composite model, wherein the cold hydrogenation process parameter composite model comprises a front-stage mechanism model, a reactor big data model and a rear-stage mechanism model; based on the initial technological parameters, the whole cold hydrogenation technological process is simulated by utilizing a cold hydrogenation technological parameter composite model to obtain predicted parameters of the cold hydrogenation technological process. Preferably, the method further comprises collecting historical process parameters and experimental analysis data of the cold hydrogenation process flow; According to the historical process parameters and experimental analysis data, an initial mechanism model of a cold hydrogenation whole process is established based on chemical thermodynamics, reaction mechanism, rectification theory and fluidized bed reaction dynamics, wherein the initial mechanism model at least comprises a heat exchanger module, a fluidized bed reactor module, a quench tower module, a cooler module, a flash tank module, a rectification tower module, a hydrogen compressor module and a pump module; carrying out correlation analysis on the historical process parameters, and screening out key parameters of the reactor, wherein the key parameters of the reactor are parameters affecting the performance of the fluidized bed reactor; According to the key parameters of the reactor in the historical process parameters, training by using a neural network to obtain a big data model of the reactor; Replacing the fluidized bed reactor module in the initial mechanism model with the reactor big data model to obtain the cold hydrogenation process parameter composite model, The front-stage mechanism model comprises the heat exchanger module, and the rear-stage mechanism model comprises the quenching tower module, the cooler module, the flash tank module, the rectifying tower module, the hydrogen compressor module and the pump module. Preferably, the correlation analysis is performed on the historical process parameters, and the screening of the key parameters of the reactor specifically includes: Carrying out pearson correlation coefficient analysis alg