A multi-parameter collaborative optimization-based industrial silicon production control system and method

Abstract

The present application relates to the technical field of production parameter control, and specifically discloses an industrial silicon production control system and method based on multi-parameter collaborative optimization, which extracts multiple-source drying distribution parameters in a monitoring area, dynamically samples the multiple-source drying distribution parameters at continuous time points, forms a multi-dimensional time sequence state sequence, and extracts the multiple-source drying change characteristics corresponding to each monitoring area; constructs a drying state analysis model according to the multiple-source drying change characteristics in the monitoring area, and obtains drying state factors in real time; screens and identifies the area where the caking risk exists according to the drying state factors, determines the risk level, evaluates the overall drying effect by comprehensively considering the drying state factors of each monitoring area, and double-determines whether the drying operation is qualified; the present application can effectively reduce the caking and blocking risks, improve the drying efficiency and reduce the energy consumption.

Description

Technical Field

[0001] This invention relates to the field of production parameter control technology, and more specifically, to an industrial silicon production control system and method based on multi-parameter collaborative optimization. Background Technology

[0002] In industrial silicon production, raw materials typically undergo drying before entering the electric furnace for smelting to reduce moisture content and improve thermal efficiency. However, due to uneven particle size distribution and significant differences in internal structure, there are often noticeable variations in moisture content between and within different particles. Furthermore, the moisture content changes dynamically during the drying process, making precise control difficult. This, coupled with the generally uneven temperature distribution within the drying equipment, further exacerbates the inconsistency in material drying, leading to some materials being over-dried or still containing high moisture content. Simultaneously, during heating, fine particulate dust and volatile organic compounds adhering to the raw materials are gradually released and migrate with the airflow, depositing, adsorbing, or even adhering within the equipment or conveying channels. Due to the combined effects of the above-mentioned multiple factors, materials that are not completely dried or have high local moisture content are prone to agglomeration during heating. The released dust and volatiles may form a highly adhesive deposition layer under the action of temperature gradient, adhering to the inner wall of the equipment or the surface of the material, further enhancing the adhesion between materials. When these agglomerated materials are transported or enter the smelting furnace material layer, they can easily cause poor material flow or even blockage, affecting the continuous and stable supply of materials.

[0003] Therefore, it is necessary to provide an industrial silicon production control system and method based on multi-parameter collaborative optimization to solve the above-mentioned technical problems. In order to solve the above problems, a technical solution is provided. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides an industrial silicon production control system and method based on multi-parameter collaborative optimization, which is used to solve the problem that material agglomeration and conveying blockage are easily caused in the existing industrial silicon drying process due to uneven moisture content, uneven temperature field and the coupling effect of dust and volatile matter deposition.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An industrial silicon production control system based on multi-parameter collaborative optimization includes a multi-source parameter acquisition module, a drying change feature extraction module, a drying status analysis module, an agglomeration risk identification module, and a drying effect discrimination module.

[0007] The multi-source parameter acquisition module is used to divide the internal monitoring area of ​​the drying equipment by deploying a multi-source sensor network inside the drying equipment, and extract the multi-source drying distribution parameters in the monitoring area respectively.

[0008] The drying change feature extraction module is used to dynamically sample the multi-source drying distribution parameters at continuous time after preprocessing the multi-source drying distribution parameters, forming a multi-dimensional time-series state sequence, and extracting the multi-source drying change features corresponding to each monitoring area.

[0009] The drying status analysis module is used to construct a drying status analysis model based on the multi-source drying change characteristics within the monitoring area and to acquire drying status factors in real time.

[0010] The agglomeration risk identification module is used to screen and identify areas where agglomeration risk exists based on drying status factors, and to determine the risk level of the areas where agglomeration risk exists.

[0011] The drying effect judgment module is used to evaluate the overall drying effect by comprehensively considering the drying status factors of each monitoring area, trigger collaborative intervention operations based on the overall drying effect, perform secondary analysis on the drying status factors after the collaborative intervention operations, and determine whether the drying operation is qualified.

[0012] As a further aspect of the present invention, the multi-source drying distribution parameters include temperature drying distribution parameters and dust drying distribution parameters; the dust drying distribution parameters include dust concentration parameters and volatile gas concentration parameters.

[0013] As a further aspect of the present invention, the drying change feature extraction module is used to dynamically sample the multi-source drying distribution parameters at continuous times after preprocessing the multi-source drying distribution parameters, forming a multi-dimensional time-series state sequence, and extracting the multi-source drying change features corresponding to each monitoring area, as follows:

[0014] The preprocessed multi-source drying distribution parameters are extracted, and the multi-source drying distribution parameters at continuous time are dynamically sampled to construct a multi-dimensional time-series state sequence. The multi-dimensional time-series state sequence includes a temperature time-series state sequence and a dust time-series state sequence.

[0015] Based on the multidimensional time-series state sequence, the multi-source drying change characteristics of each monitoring area are extracted. The multi-source drying change characteristics include temperature drying change values ​​and dust drying change values; the dust drying change values ​​include dust concentration change values ​​and volatile gas concentration change values.

[0016] As a further aspect of the present invention, the drying status analysis module constructs a drying status analysis model based on the multi-source drying change characteristics within the monitoring area, and acquires drying status factors in real time. These drying status factors include drying temperature status factors and drying dust status factors, as detailed below:

[0017] By extracting the temperature drying change value from the multi-source drying change characteristics, a first-level drying state analysis model is constructed, and the drying temperature state factor is calculated.

[0018] By extracting the dust drying change value from the multi-source drying change characteristics, a two-level drying state analysis model is constructed to calculate the drying dust state factor; the drying dust state factor includes the first dust state factor and the second dust state factor.

[0019] As a further aspect of the present invention, the agglomeration risk identification module identifies areas where agglomeration risk exists based on drying status factors, and determines the risk level of the areas where agglomeration risk exists. The specific steps are as follows:

[0020] A coupling identification strategy for agglomeration risk is established based on drying temperature state factors and drying dust state factors to screen and identify areas where agglomeration risk exists.

[0021] For areas where agglomeration risk exists, the corresponding drying status factors are extracted and weighted to obtain a comprehensive agglomeration risk index. Based on the comprehensive agglomeration risk index, the risk level of the areas where agglomeration risk exists is determined.

[0022] As a further aspect of the present invention, an agglomeration risk coupling identification strategy is established based on the drying temperature state factor and the drying dust state factor to screen and identify areas where agglomeration risk exists. Specifically, an agglomeration risk coupling identification strategy is established based on the drying temperature state factor and the drying dust state factor to sequentially mark the areas where agglomeration risk exists to be identified. If a monitoring area is marked as an area where agglomeration risk exists three times, then the corresponding monitoring area is identified as an area where agglomeration risk exists.

[0023] As a further aspect of the present invention, the multi-blocking risk coupling identification strategy includes a primary blocking risk identification strategy, a secondary blocking risk identification strategy, and a tertiary blocking risk identification strategy.

[0024] The specific strategy for identifying agglomeration risk is as follows: compare the drying temperature state factor with the preset temperature drying threshold. If the drying temperature state factor is greater than or equal to the preset temperature drying threshold, the corresponding monitoring area is marked as a non-agglomeration risk area; if the drying temperature state factor is less than the preset temperature drying threshold, the corresponding monitoring area is marked as an area where agglomeration risk exists.

[0025] The secondary agglomeration risk identification strategy is as follows: the first dust state factor is compared with the preset first dust drying threshold. If the first dust state factor is greater than or equal to the preset first dust drying threshold, the corresponding monitoring area is marked as a non-agglomeration risk area; if the first dust state factor is less than the preset first dust drying threshold, the corresponding monitoring area is marked as an area where agglomeration risk exists.

[0026] The three-stage agglomeration risk identification strategy is as follows: the second dust state factor is compared with the preset second dust drying threshold. If the second dust state factor is greater than or equal to the preset second dust drying threshold, the corresponding monitoring area is marked as a non-agglomeration risk area; if the second dust state factor is less than the preset second dust drying threshold, the corresponding monitoring area is marked as an area where agglomeration risk exists.

[0027] As a further aspect of the present invention, the drying effect judgment module evaluates the overall drying effect by comprehensively considering the drying status factors of each monitoring area, triggers a collaborative intervention operation based on the overall drying effect, performs a secondary analysis of the drying status factors after the collaborative intervention operation, and determines whether the drying operation is qualified, as follows:

[0028] By summarizing the drying status factors of each area with clumping risk in real time, the drying status effect coefficient is obtained, and the overall drying effect is evaluated based on the drying status effect coefficient. Specifically, the drying status effect coefficient is compared with the drying effect threshold. If the drying status effect coefficient is greater than or equal to the drying effect threshold, the drying operation is unqualified. If the drying status effect coefficient is less than the drying effect threshold, the drying operation is qualified and drying continues.

[0029] When the drying effect is unqualified, a collaborative intervention operation is triggered. The drying status factors after the collaborative intervention operation are analyzed again to determine whether the drying operation is qualified. An alarm reminder is triggered for unqualified drying operations.

[0030] An industrial silicon production control method based on multi-parameter collaborative optimization, the specific steps of which are as follows:

[0031] By deploying a multi-source sensor network inside the drying equipment, the internal monitoring area of ​​the drying equipment is divided, and the multi-source drying distribution parameters within the monitoring area are extracted respectively.

[0032] Based on the preprocessing of the multi-source drying distribution parameters, the multi-source drying distribution parameters at continuous time are dynamically sampled to form a multi-dimensional time-series state sequence, and the multi-source drying change characteristics corresponding to each monitoring area are extracted respectively.

[0033] A drying status analysis model is constructed based on the multi-source drying change characteristics within the monitoring area to obtain drying status factors in real time.

[0034] Based on drying status factors, areas with agglomeration risk are identified, and the risk level of these areas is determined.

[0035] The overall drying effect is evaluated by comprehensively considering the drying status factors of each monitoring area. Based on the overall drying effect, a collaborative intervention operation is triggered. The drying status factors after the collaborative intervention operation are analyzed again to determine whether the drying operation is qualified.

[0036] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of an industrial silicon production control system and method based on multi-parameter collaborative optimization as described above.

[0037] The technical effects and advantages of this invention, a multi-parameter collaborative optimization-based industrial silicon production control system and method, are as follows: The multi-source parameter acquisition module accurately captures local anomalies such as localized high moisture content or dust deposition, avoiding misjudgments due to incomplete information and improving data reliability from the source. The drying change feature extraction module not only reflects the current state but also reveals the evolutionary patterns of temperature and dust, providing more discriminative input for subsequent analysis and enhancing sensitivity and foresight regarding abnormal operating conditions. The drying state analysis module not only improves the consistency of judgments but also facilitates subsequent automated control and model optimization. The agglomeration risk identification module effectively reduces the misjudgment rate through multiple rounds of screening, accurately locating and classifying agglomeration risk areas, providing a basis for differentiated control and avoiding a one-size-fits-all, crude control approach. The drying effect judgment module not only judges whether drying is qualified in real time but also automatically triggers control and verifies the effect when unqualified, realizing a shift from passive monitoring to proactive optimization and improving the system's adaptability and operational stability.

[0038] This invention not only enables real-time monitoring of the complex multi-field coupling state inside the drying equipment, but also allows for early identification and proactive control of agglomeration risks, effectively preventing equipment blockage caused by material agglomeration. Furthermore, through continuous feedback and optimization, it gradually improves the system's adaptability to different raw material characteristics and operating conditions, thereby significantly enhancing the stability, continuity, and energy efficiency of the industrial silicon production process. Attached Figure Description

[0039] Figure 1 A schematic diagram of the structure of an industrial silicon production control system based on multi-parameter collaborative optimization provided in an embodiment of the present invention;

[0040] Figure 2 This is a schematic diagram of a process for industrial silicon production control based on multi-parameter collaborative optimization, provided as an embodiment of the present invention. Detailed Implementation

[0041] The technical solutions of this invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described technical solutions are only a part of this invention, and not all of it. All other technical solutions obtained by those skilled in the art based on the technical solutions of this invention without inventive effort are within the scope of protection of this invention.

[0042] Example 1: As Figure 1 The diagram shown is a structural schematic of an industrial silicon production control system based on multi-parameter collaborative optimization provided by an embodiment of the present invention. The industrial silicon production control system based on multi-parameter collaborative optimization includes a multi-source parameter acquisition module, a drying change feature extraction module, a drying state analysis module, an agglomeration risk identification module, and a drying effect discrimination module. The multi-source parameter acquisition module is connected to the drying change feature extraction module, the drying change feature extraction module is connected to the drying state analysis module, the drying state analysis module is connected to the agglomeration risk identification module, and the agglomeration risk identification module is connected to the drying effect discrimination module.

[0043] The multi-source parameter acquisition module is used to divide the internal monitoring area of ​​the drying equipment by deploying a multi-source sensor network inside the drying equipment, and extract the multi-source drying distribution parameters in the monitoring area respectively.

[0044] The drying change feature extraction module is used to dynamically sample the multi-source drying distribution parameters at continuous time after preprocessing the multi-source drying distribution parameters, forming a multi-dimensional time-series state sequence, and extracting the multi-source drying change features corresponding to each monitoring area.

[0045] The drying status analysis module is used to construct a drying status analysis model based on the multi-source drying change characteristics within the monitoring area and to acquire drying status factors in real time.

[0046] The agglomeration risk identification module is used to screen and identify areas where agglomeration risk exists based on drying status factors, and to determine the risk level of the areas where agglomeration risk exists.

[0047] The drying effect judgment module is used to evaluate the overall drying effect by comprehensively considering the drying status factors of each monitoring area, trigger collaborative intervention operations based on the overall drying effect, perform secondary analysis on the drying status factors after the collaborative intervention operations, and determine whether the drying operation is qualified.

[0048] It should be noted that all parameter data involved in the calculation were dimensionless before the calculation was performed.

[0049] Preferably, the multi-source drying distribution parameters include temperature drying distribution parameters and dust drying distribution parameters; the dust drying distribution parameters include dust concentration parameters and volatile gas concentration parameters.

[0050] Preferably, the drying change feature extraction module is used to dynamically sample the multi-source drying distribution parameters at continuous time intervals after preprocessing them, forming a multi-dimensional time-series state sequence, and extracting the multi-source drying change features corresponding to each monitoring area, as follows:

[0051] Extract the preprocessed multi-source drying distribution parameters, dynamically sample the multi-source drying distribution parameters at continuous time points, and construct a multi-dimensional time-series state sequence; the multi-dimensional time-series state sequence includes a temperature time-series state sequence. and dust time-series state sequence ,in, The location of the monitoring area is Temperature drying distribution parameters at time i The location of the monitoring area is Dust drying distribution parameters at time i For monitoring duration;

[0052] Based on the multidimensional time-series state sequence, the multi-source drying change characteristics of each monitoring area are extracted. The multi-source drying change characteristics include temperature drying change values ​​and dust drying change values; the dust drying change values ​​include dust concentration change values ​​and volatile gas concentration change values.

[0053] More specifically, the acquired multi-source drying distribution parameters undergo unified preprocessing. For temperature and dust drying distribution parameters, outlier removal and noise filtering are performed respectively. Sliding window mean or median filtering is used to eliminate instantaneous fluctuations, while threshold judgment is used to remove abrupt outliers. Based on this, parameters of different dimensions are normalized to map each parameter to a unified numerical range, ensuring the comparability of subsequent multi-parameter fusion analysis. It should be noted that the preprocessed multi-source drying distribution parameters undergo time alignment and spatial matching. To address the issue of inconsistent sampling frequencies from different sensors, a unified time step is used as a benchmark. High-frequency data is downsampled, and low-frequency data is interpolated to compensate for inconsistencies, constructing a time synchronization sequence. Simultaneously, based on the monitoring area division within the drying equipment, data from different sensors are mapped to corresponding monitoring areas, forming a data set based on "monitoring area," achieving consistent spatial representation.

[0054] In one embodiment of the present invention, during the drying process in industrial silicon production, multi-source sensor nodes can be arranged axially and radially inside a rotary kiln or fluidized bed drying equipment to divide the equipment into several monitoring areas. Within each monitoring area, temperature sensors collect real-time temperature drying distribution parameters of the material and the gaseous environment, while a dust monitoring unit simultaneously acquires dust concentration and volatile gas concentration parameters, forming multi-source drying distribution parameters. First, the collected data undergoes preprocessing, including noise reduction, outlier removal, and normalization, to eliminate the influence of sensing errors and different dimensions. Based on this, data from consecutive moments are dynamically sampled at fixed time intervals, and the temperature and dust changes in each monitoring area throughout the entire monitoring period are expressed temporally, thereby constructing corresponding multi-dimensional temporal state sequences, namely, temperature temporal state sequences and dust temporal state sequences.

[0055] In specific operations, for example, if a monitoring area is located in the middle of a region with a thick accumulation of material, its temperature time series may show a slow rise and small fluctuations, while the dust time series may show a trend of first rising and then falling, corresponding to the process of dust deposition after release. Based on the multi-dimensional time series, the multi-source drying change characteristics of this area are further extracted. The temperature drying change value can be characterized by the temperature difference or temperature change rate between adjacent time points, reflecting the drying rate and heat transfer efficiency of this area; the dust drying change value is characterized by the dust concentration change value and the volatile gas concentration change value, reflecting the dust release intensity and volatilization behavior of the material during heating.

[0056] In actual operating conditions, when the temperature change in a certain area is low and lasts for a long time, while the dust concentration change shows a trend from high to low, and the volatile gas concentration change shows periodic peaks, it can be determined that the area contains high-moisture materials accompanied by dust deposition, which is a typical precursor area to agglomeration risk. Conversely, for areas with high and stable temperature changes and small dust changes, it indicates that the drying process in that area is uniform and stable. This embodiment of the invention, through the extraction and analysis of the above-mentioned multi-source drying change characteristics, can achieve refined identification of the differences in drying status in different monitoring areas during actual production, providing a reliable data foundation for subsequent agglomeration risk assessment and coordinated control.

[0057] Preferably, the drying status analysis module constructs a drying status analysis model based on the multi-source drying change characteristics within the monitoring area, and acquires drying status factors in real time. These drying status factors include drying temperature status factors and drying dust status factors, as detailed below:

[0058] By extracting temperature drying variation values ​​from the multi-source drying variation characteristics, a primary drying state analysis model is constructed to calculate the drying temperature state factor; the calculation formula for the primary drying state analysis model is as follows:

[0059]

[0060] In the formula: The location of the monitoring area is The drying temperature state factor, The location of the monitoring area is Temperature drying distribution parameters at time i+1 The location of the monitoring area is The temperature drying change at time i;

[0061] By extracting dust drying variation values ​​from the multi-source drying variation characteristics, a two-level drying state analysis model is constructed to calculate the drying dust state factor. The drying dust state factor includes a first dust state factor and a second dust state factor. The calculation formula for the two-level drying state analysis model is as follows:

[0062]

[0063] In the formula: The location of the monitoring area is The first dust state factor, The location of the monitoring area is The dust concentration parameter at time i+1 The location of the monitoring area is The dust concentration parameter at time i. The location of the monitoring area is The change in dust concentration at time i;

[0064]

[0065] In the formula: The location of the monitoring area is The second dust state factor, The location of the monitoring area is The volatile gas concentration parameters at time i+1 The location of the monitoring area is The volatile gas concentration parameters at time i. The location of the monitoring area is , the change in volatile gas concentration at time i.

[0066] In one embodiment of the present invention, in the drying process of industrial silicon production, a continuously operating rotary kiln can be divided into multiple monitoring zones along the axial and radial directions. Temperature sensors, dust concentration sensors, and volatile gas detection units are installed in each zone. During operation, temperature drying distribution parameters and dust and volatile gas concentration parameters are continuously collected from each monitoring zone at different time points. The data from consecutive moments are processed based on a sliding time window to extract temperature drying variation values ​​and dust drying variation values. Subsequently, a hierarchical analysis model is constructed based on the aforementioned variation characteristics to quantitatively evaluate the drying status of each zone.

[0067] Specifically, in a certain monitoring area, such as the middle section where the material is thickly piled, the temperature change value between adjacent time periods is first calculated based on the temperature data of continuous time. Then, the drying temperature state factor of the area is obtained through the first-level drying state analysis model, which reflects the average amplitude of temperature change per unit time. When the value is small, it indicates that the temperature change is slow, which usually corresponds to the state of high material moisture content and obstructed drying process. When the value is large, it indicates that the temperature change in the area is active, the heat transfer efficiency is high, and the material is dried more thoroughly.

[0068] Meanwhile, based on the continuous changes in dust concentration and volatile gas concentration, the first and second dust state factors were calculated using a two-stage drying state analysis model. In real-world scenarios, for example, when a large amount of fine dust is released during the material's heating process, the dust concentration will fluctuate significantly. In this case, a larger first dust state factor indicates strong dust activity in the area. Conversely, when the dust concentration changes gradually or even decreases, it may mean that dust has deposited or adhered. Similarly, changes in volatile gas concentration reflect the volatilization process of moisture and organic components in the material. A larger second dust state factor indicates a vigorous volatilization process, typically corresponding to a high moisture content or rapid drying stage.

[0069] By comprehensively analyzing the drying temperature state factor and two types of dust state factors, the drying state of different monitoring areas can be clearly characterized in actual production. For example, when a certain area exhibits a low temperature state factor but high dust and volatile gas state factors, it can be determined that the area is in a stage of high moisture content, strong volatility, and active dust, which is highly prone to dust adhesion and agglomeration. Conversely, when the temperature state factor is high and the dust state factor is low, it indicates that the area is drying uniformly and operating stably. This embodiment of the invention, through a drying state analysis model constructed based on multi-source variation characteristics, can achieve refined quantification and dynamic monitoring of the drying state in actual industrial silicon production, providing a reliable basis for subsequent agglomeration risk identification and coordinated control.

[0070] Preferably, the agglomeration risk identification module identifies areas with agglomeration risk based on drying status factors and determines the risk level of these areas. The specific steps are as follows:

[0071] A coupling identification strategy for agglomeration risk is established based on drying temperature state factors and drying dust state factors to screen and identify areas where agglomeration risk exists.

[0072] For areas where agglomeration risk exists, the corresponding drying status factors are extracted and weighted to obtain a comprehensive agglomeration risk index. Based on the comprehensive agglomeration risk index, the risk level of the areas where agglomeration risk exists is determined.

[0073] Preferably, an agglomeration risk coupling identification strategy is established based on the drying temperature state factor and the drying dust state factor to screen and identify areas where agglomeration risk exists. Specifically, a multiple agglomeration risk coupling identification strategy is established based on the drying temperature state factor and the drying dust state factor, and the areas where agglomeration risk exists to be identified are marked sequentially. If a monitoring area is marked as an area where agglomeration risk exists three times, then the corresponding monitoring area is identified as an area where agglomeration risk exists. The multiple agglomeration risk coupling identification strategy includes a primary agglomeration risk identification strategy, a secondary agglomeration risk identification strategy, and a tertiary agglomeration risk identification strategy.

[0074] Preferably, the primary agglomeration risk identification strategy is as follows: The drying temperature state factor and the drying dust state factor are compared with preset drying thresholds to identify areas where agglomeration risk exists. Specifically, the drying temperature state factor is compared with the preset temperature drying threshold. If the drying temperature state factor is greater than or equal to the preset temperature drying threshold, the corresponding monitoring area is marked as a non-agglomeration risk area; if the drying temperature state factor is less than the preset temperature drying threshold, the corresponding monitoring area is marked as an area where agglomeration risk exists to be identified.

[0075] The secondary agglomeration risk identification strategy is as follows: the first dust state factor is compared with the preset first dust drying threshold. If the first dust state factor is greater than or equal to the preset first dust drying threshold, the corresponding monitoring area is marked as a non-agglomeration risk area; if the first dust state factor is less than the preset first dust drying threshold, the corresponding monitoring area is marked as an area where agglomeration risk exists.

[0076] The three-stage agglomeration risk identification strategy is as follows: the second dust state factor is compared with the preset second dust drying threshold. If the second dust state factor is greater than or equal to the preset second dust drying threshold, the corresponding monitoring area is marked as a non-agglomeration risk area; if the second dust state factor is less than the preset second dust drying threshold, the corresponding monitoring area is marked as an area where agglomeration risk exists.

[0077] In an embodiment of the present invention, during the actual drying process of industrial silicon production, when a rotary kiln or large fluidized bed equipment is running, the interior of the equipment can be divided into multiple monitoring zones. The drying temperature state factors and two types of dust state factors for each zone are acquired in real time using the aforementioned drying state analysis model. An agglomeration risk identification module operates on this basis, constructing a multi-stage agglomeration risk coupling identification strategy to perform layer-by-layer screening and discrimination of each monitoring zone, thereby improving the accuracy and stability of agglomeration identification under complex operating conditions.

[0078] In actual operation, the agglomeration risk identification strategy is executed first. For example, in a monitoring area located in the feeding section, if its drying temperature state factor is significantly lower than the preset temperature drying threshold, it indicates that the temperature change in this area is slow and the heat transfer is insufficient, which usually corresponds to a high material moisture content. In this case, the area is marked as an area where agglomeration risk exists. Conversely, for areas with active temperature changes and a drying temperature state factor higher than the threshold, they are directly determined as non-agglomeration risk areas, thus achieving the first round of screening and quickly eliminating most of the sufficiently dried and stable areas.

[0079] Subsequently, a secondary agglomeration risk identification strategy is implemented for the areas that passed the first screening. In this stage, the focus is on analyzing the characteristics of dust concentration changes. For example, in a certain central region, if the first dust state factor is low, it indicates that the dust concentration change is small, often meaning that dust has already settled from the gas phase and adhered to the material surface. In this case, the area is still marked as an area with agglomeration risk. Conversely, for areas with active dust changes and a high first dust state factor, it indicates that the dust is still in a suspended state and is unlikely to form an adhesive structure. These areas can be identified as non-agglomeration risk areas, thus completing the second round of screening.

[0080] Based on this, a three-stage agglomeration risk identification strategy is further implemented to analyze changes in volatile gas concentrations. For example, in a certain pre-discharge area, if the second dust state factor is low, it indicates that the volatilization process is stabilizing or weakening, which usually corresponds to the near completion of moisture evaporation and the presence of local condensation conditions. Such areas are more prone to forming wet sticky layers and dust deposition agglomeration structures, and therefore are marked as areas where agglomeration risk exists. Conversely, if the volatile gas changes drastically, it indicates that the area is still in an active drying stage and agglomeration is unlikely to occur, so it can be identified as a non-risk area.

[0081] When a monitoring area is marked as an "area with potential for agglomeration risk" in all three identification processes mentioned above, the area is ultimately confirmed as an area with potential agglomeration risk. For example, in actual operation, if an area simultaneously exhibits characteristics of slow temperature change, small dust fluctuation, and reduced volatilization, then the area is very likely to have entered the agglomeration formation stage.

[0082] After identifying the agglomeration risk areas, the drying state factors of the corresponding areas are further extracted, and a comprehensive agglomeration risk index is constructed through weighted fusion to comprehensively reflect the coupled influence of temperature, dust, and volatilization behavior. In practical applications, for example, areas near material accumulation dead corners usually have a higher comprehensive agglomeration risk index and can be identified as high-risk areas; while edge transition areas may be identified as medium-risk areas. This hierarchical judgment method can not only identify the location of agglomeration but also clarify its risk level, thus providing a precise basis for subsequent differentiated control.

[0083] Preferably, the drying effect judgment module evaluates the overall drying effect by comprehensively considering the drying status factors of each monitoring area, triggers a collaborative intervention operation based on the overall drying effect, performs a secondary analysis of the drying status factors after the collaborative intervention operation, and determines whether the drying operation is qualified, as follows:

[0084] By summarizing the drying status factors of each area with clumping risk in real time, the drying status effect coefficient is obtained, and the overall drying effect is evaluated based on the drying status effect coefficient. Specifically, the drying status effect coefficient is compared with the drying effect threshold. If the drying status effect coefficient is greater than or equal to the drying effect threshold, the drying operation is unqualified. If the drying status effect coefficient is less than the drying effect threshold, the drying operation is qualified and drying continues.

[0085] When the drying effect is unsatisfactory, a collaborative intervention operation is triggered. A second analysis of the drying status factors after the intervention is performed to reassess whether the drying operation is satisfactory. An alarm is triggered for any unsatisfactory drying operation. The collaborative intervention operation includes the linkage of temperature control and dust control. For example, adjusting the hot air distribution to improve temperature uniformity, while optimizing airflow speed and direction to reduce the risk of dust deposition, can be combined with feed rate adjustment to achieve multi-parameter collaborative optimization.

[0086] In an embodiment of the present invention, during the actual drying process of industrial silicon production, a continuously operating rotary kiln system has been divided into several monitoring areas through a multi-source sensor network, and the drying temperature state factor and dust state factor of each area are acquired in real time. After completing the identification of agglomeration risk, the data of areas identified as having agglomeration risk are summarized and processed. By weighted and fused drying state factors of these key areas, a drying state effect coefficient that can characterize the overall drying quality is formed, reflecting the comprehensive degree of insufficient temperature uniformity and dust deposition trend in the current drying process.

[0087] In specific operational scenarios, if the calculated drying effect coefficient remains consistently higher than the preset drying effect threshold after the equipment has been running for a period of time, it indicates that several key areas still suffer from slow temperature changes and significant dust accumulation, and the overall drying effect fails to meet the expected standard, thus classifying the current drying operation as unqualified. At this point, the drying effect judgment module automatically triggers collaborative intervention to adjust the equipment operating parameters in a coordinated manner. For example, in temperature control, adjusting the hot air distribution ratio in different areas or increasing local heating power improves heat supply to low-temperature areas and reduces temperature unevenness; in dust control, increasing airflow velocity, optimizing airflow direction, or introducing pulse disturbances reduces dust deposition in high-moisture areas; simultaneously, the feed rate can be appropriately reduced to allow the material more sufficient drying time, thereby achieving coordinated optimization control of temperature and dust.

[0088] After completing one round of collaborative intervention, real-time monitoring of each monitoring area continues, and the drying state factor after intervention is recalculated. By comparing the state changes before and after intervention, such as whether the temperature state factor has increased or whether the dust state factor has stabilized, the drying state effect coefficient is recalculated and compared with the drying effect threshold again. In practical applications, if the drying state effect coefficient decreases significantly after intervention and is lower than the threshold, it indicates that the control measures are effective, the drying process has returned to a qualified state, and the current operating strategy should be maintained or fine-tuned and optimized. If the coefficient is still higher than the threshold, it indicates that the local agglomeration or deposition problem has not been resolved, the drying operation is still deemed unqualified, and an alarm mechanism is further triggered to prompt the operator or the upper control system to strengthen intervention, such as increasing the control intensity or switching the operating mode. In real industrial silicon production scenarios, this invention can realize a complete control link from local risk identification to global effect evaluation, and then to collaborative intervention and closed-loop verification. This not only improves the stability of the drying process but also effectively reduces the risk of blockage caused by agglomeration, ensuring the continuity and safety of production.

[0089] Example 2: In one embodiment of the present invention, an industrial silicon production control system based on multi-parameter collaborative optimization is applied to the industrial silicon production process. The drying process typically relies on large thermal equipment such as rotary kilns or fluidized beds to remove moisture from the raw materials. However, due to the uneven particle size distribution, large differences in moisture content, and complex gas-solid two-phase flow of the raw materials, traditional control methods relying on experience or single-point monitoring are difficult to detect local anomalies in a timely manner, easily leading to material agglomeration or even equipment blockage. Therefore, in a practical production scenario, an industrial silicon production control system based on multi-parameter collaborative optimization is introduced, consisting of a multi-source parameter acquisition module, a drying change feature extraction module, a drying state analysis module, an agglomeration risk identification module, and a drying effect discrimination module, to achieve intelligent management of the entire drying process.

[0090] Specifically, within the rotary kiln equipment of an industrial silicon production line, a multi-source parameter acquisition module is first deployed. This module divides the drying space into several discrete monitoring areas—such as annular sections and radially layered areas—by arranging various types of sensor nodes along the axial and radial directions within the kiln body. Each monitoring area can collect real-time temperature drying distribution parameters, dust concentration parameters, and volatile gas concentration parameters, thus forming a multi-source drying distribution parameter field covering the entire equipment. In actual operation, for example, the feed end area often contains materials with high moisture content, resulting in lower temperatures and higher volatile gas concentrations, while the middle and later sections may experience dust accumulation or deposition. The aforementioned multi-source data provides a comprehensive initial sensing basis.

[0091] After acquiring the raw data, the drying change feature extraction module preprocesses the multi-source drying distribution parameters, including removing outliers, filtering high-frequency noise, and normalizing data of different dimensions, thus ensuring the stability and consistency of subsequent analysis. Based on a sliding time window mechanism, multi-source parameters at continuous time points are dynamically sampled to construct a multi-dimensional time-series state sequence. For example, for a certain monitoring area, its temperature time-series state sequence can reflect the temperature rise process of the area over a period of time, while the dust and volatile gas time-series sequences reflect the dynamic changes in particle release and moisture evaporation. Furthermore, by calculating the difference or rate of change of data between adjacent time points, the temperature drying change value and the dust drying change value are extracted. In real-world scenarios, for example, a central region may exhibit slow temperature changes while dust concentration first rises and then falls; this change feature can be accurately extracted and used for subsequent analysis.

[0092] After obtaining the multi-source drying variation characteristics, state modeling is performed on each monitoring area. By constructing a drying state analysis model, temperature change characteristics are mapped to drying temperature state factors, and dust concentration changes and volatile gas changes are mapped to a first dust state factor and a second dust state factor, respectively, thereby achieving a quantitative expression of the drying state of each monitoring area. In actual operation, for example, if the drying temperature state factor is low in a certain area, it indicates that the temperature rise in that area is slow and there may be high moisture content materials; if the first dust state factor is low, it may mean that dust has been deposited on the surface of the material; if the second dust state factor is low, it indicates that the volatilization process is weakened or even that there is a condensation trend. Through comprehensive analysis of the three types of state factors, the degree of dryness, temperature uniformity, and dust activity in different areas can be accurately characterized.

[0093] Based on the aforementioned state factors, each monitoring area is screened and identified. A multi-stage agglomeration risk coupling identification strategy is constructed, jointly analyzing temperature state factors and dust state factors. For example, in the first identification process, if the temperature state factor of a certain area is below the threshold, it is initially identified as a potentially high-moisture area. In the second identification, if the dust state factor of the same area is also low, it further indicates that dust deposition has occurred. In the third identification, if the volatile gas state factor is also low, it indicates that evaporation in the area is weakened and condensation conditions may exist. When a monitoring area is marked as abnormal three times consecutively, it is finally identified as an area with agglomeration risk. In actual production, such areas often appear where material accumulation is thick or airflow disturbance is insufficient. Furthermore, by weighted fusion of the state factors of this area, a comprehensive agglomeration risk index is constructed, and the risk is divided into low-risk, medium-risk, and high-risk levels according to the index magnitude, thereby achieving refined risk classification management.

[0094] After identifying the risk of agglomeration, a global assessment of the entire drying process is performed. Drying state factors for all areas with agglomeration risk are summarized and weighted to form a drying state effect coefficient, which characterizes the overall quality of the current drying process. In actual operation, for example, when multiple key areas simultaneously exhibit slow temperature changes and increased dust deposition, the calculated drying state effect coefficient is often high. In this case, it is compared with a preset drying effect threshold, and the current drying operation is determined to be unqualified. In response, a coordinated intervention operation is automatically triggered, optimizing control through coordinated adjustment of temperature and airflow parameters. For example, adjusting the hot air distribution increases the temperature in low-temperature areas, improving temperature uniformity; increasing local airflow velocity or changing its direction reduces dust deposition; and combining this with feed rate adjustment makes the material's residence time in the equipment more reasonable, thereby achieving multi-parameter coordinated optimization.

[0095] After the intervention is implemented, a secondary analysis phase is initiated. Data is collected and state factors are calculated again for each monitoring area, and a new drying state effect coefficient is generated and compared with the threshold to determine the intervention effect. In a real-world scenario, if the temperature state factor significantly improves and the dust state factor stabilizes after the intervention, it indicates that the control is effective, and the drying operation is deemed to have recovered satisfactorily. Conversely, if the state improvement is not significant, it indicates that the problem has not been resolved. In this case, the control strategy will be readjusted, and an alarm mechanism will be triggered to remind operators to conduct further intervention or check the equipment's operating status.

[0096] This invention not only enables real-time monitoring of the complex multi-field coupling state inside the drying equipment, but also allows for early identification and proactive control of agglomeration risks, effectively preventing equipment blockage caused by material agglomeration. Furthermore, through continuous feedback and optimization, it can gradually improve adaptability to different raw material characteristics and operating conditions, thereby significantly enhancing the stability, continuity, and energy efficiency of the industrial silicon production process.

[0097] Example 3: An industrial silicon production control method based on multi-parameter collaborative optimization, the specific steps of which are as follows:

[0098] By deploying a multi-source sensor network inside the drying equipment, the internal monitoring area of ​​the drying equipment is divided, and multi-source drying distribution parameters are extracted in each monitoring area. The multi-source drying distribution parameters include temperature drying distribution parameters and dust drying distribution parameters. The dust drying distribution parameters include dust concentration parameters and volatile gas concentration parameters.

[0099] Based on the preprocessing of the multi-source drying distribution parameters, the multi-source drying distribution parameters at continuous time are dynamically sampled to form a multi-dimensional time-series state sequence, and the multi-source drying change characteristics corresponding to each monitoring area are extracted respectively.

[0100] A drying status analysis model is constructed based on the multi-source drying change characteristics within the monitoring area to obtain drying status factors in real time.

[0101] Based on drying status factors, areas with agglomeration risk are identified, and the risk level of these areas is determined.

[0102] The overall drying effect is evaluated by comprehensively considering the drying status factors of each monitoring area. Based on the overall drying effect, a collaborative intervention operation is triggered. The drying status factors after the collaborative intervention operation are analyzed again to determine whether the drying operation is qualified.

[0103] like Figure 2 The diagram shown is a flowchart of an industrial silicon production control method based on multi-parameter collaborative optimization according to an embodiment of the present invention, which can be used to execute... Figure 1 The steps in the method embodiments shown are implemented in a similar manner and have similar technical effects, and will not be repeated here.

[0104] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of an industrial silicon production control system based on multi-parameter collaborative optimization as described above.

[0105] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0106] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0107] In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0108] Through the above embodiments, this invention achieves comprehensive perception and spatial visualization monitoring of the drying process by finely dividing the interior of the drying equipment and introducing multi-source information such as temperature, dust, and volatile gases. Compared with traditional single-point or single-parameter monitoring methods, this module can accurately capture local anomalies, such as local high moisture content or dust deposition, avoiding misjudgments caused by incomplete information and improving data reliability from the source. By constructing a multi-dimensional time-series state sequence, the originally discrete and static data is transformed into continuous dynamic change characteristics, realizing the characterization of the "change trend" of the drying process. Its advantage is that it can not only reflect the current state but also reveal the evolution law of temperature and dust, thereby providing more discriminative input for subsequent analysis and improving the system's sensitivity and foresight to abnormal operating conditions. By constructing a state factor model, the complex multi-source change characteristics are compressed into quantifiable drying temperature state factors and dust state factors, realizing a standardized expression of the drying state. This not only improves the consistency of judgment but also facilitates subsequent automated control and model optimization. By employing a multi-coupling identification strategy, the accuracy and robustness of agglomeration identification are significantly improved. Compared to single-threshold judgment, multi-round screening effectively reduces the false judgment rate, accurately locates and classifies agglomeration risk areas, and provides a basis for differentiated regulation, thus avoiding a one-size-fits-all, crude control approach. By comprehensively evaluating global state factors, an overall drying effect judgment mechanism is constructed. Combined with collaborative intervention and secondary verification, a closed-loop control is formed. This not only determines whether drying is qualified in real time but also automatically triggers regulation and verifies the effect when unqualified, realizing a shift from passive monitoring to proactive optimization and significantly improving the system's adaptability and operational stability. This invention achieves closed-loop intelligent control from local perception to global decision-making, and from state recognition to predictive intervention, effectively reducing the risk of agglomeration and blockage, improving drying efficiency, and reducing energy consumption.

[0109] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

[0110] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An industrial silicon production control system based on multi-parameter collaborative optimization, characterized in that, It includes a multi-source parameter acquisition module, a drying change feature extraction module, a drying status analysis module, an agglomeration risk identification module, and a drying effect discrimination module; The multi-source parameter acquisition module is used to divide the internal monitoring area of ​​the drying equipment by deploying a multi-source sensor network inside the drying equipment, and extract the multi-source drying distribution parameters in the monitoring area respectively. The drying change feature extraction module is used to dynamically sample the multi-source drying distribution parameters at continuous time after preprocessing the multi-source drying distribution parameters, forming a multi-dimensional time-series state sequence, and extracting the multi-source drying change features corresponding to each monitoring area. The drying status analysis module is used to construct a drying status analysis model based on the multi-source drying change characteristics within the monitoring area and to acquire drying status factors in real time. The agglomeration risk identification module is used to screen and identify areas where agglomeration risk exists based on drying status factors, and to determine the risk level of the areas where agglomeration risk exists. The drying effect judgment module is used to evaluate the overall drying effect by comprehensively considering the drying status factors of each monitoring area, trigger collaborative intervention operations based on the overall drying effect, perform secondary analysis on the drying status factors after the collaborative intervention operations, and determine whether the drying operation is qualified.

2. The industrial silicon production control system based on multi-parameter collaborative optimization according to claim 1, characterized in that, Multi-source drying distribution parameters include temperature drying distribution parameters and dust drying distribution parameters; dust drying distribution parameters include dust concentration parameters and volatile gas concentration parameters.

3. The industrial silicon production control system based on multi-parameter collaborative optimization according to claim 1, characterized in that, The drying change feature extraction module is used to dynamically sample the multi-source drying distribution parameters at continuous time intervals after preprocessing them, forming a multi-dimensional time-series state sequence, and extracting the multi-source drying change features corresponding to each monitoring area, as follows: The preprocessed multi-source drying distribution parameters are extracted, and the multi-source drying distribution parameters at continuous time are dynamically sampled to construct a multi-dimensional time-series state sequence. The multi-dimensional time-series state sequence includes a temperature time-series state sequence and a dust time-series state sequence. Based on the multidimensional time-series state sequence, the multi-source drying change characteristics of each monitoring area are extracted. The multi-source drying change characteristics include temperature drying change values ​​and dust drying change values; the dust drying change values ​​include dust concentration change values ​​and volatile gas concentration change values.

4. The industrial silicon production control system based on multi-parameter collaborative optimization according to claim 1, characterized in that, The drying status analysis module constructs a drying status analysis model based on the multi-source drying change characteristics within the monitoring area, and acquires drying status factors in real time. These drying status factors include drying temperature status factors and drying dust status factors, as detailed below: By extracting the temperature drying change value from the multi-source drying change characteristics, a first-level drying state analysis model is constructed, and the drying temperature state factor is calculated. By extracting the dust drying change value from the multi-source drying change characteristics, a two-level drying state analysis model is constructed to calculate the drying dust state factor; the drying dust state factor includes the first dust state factor and the second dust state factor.

5. The industrial silicon production control system based on multi-parameter collaborative optimization according to claim 4, characterized in that, The agglomeration risk identification module identifies areas with agglomeration risk based on drying status factors and determines the risk level of these areas. The specific steps are as follows: A coupling identification strategy for agglomeration risk is established based on drying temperature state factors and drying dust state factors to screen and identify areas where agglomeration risk exists. For areas where agglomeration risk exists, the corresponding drying status factors are extracted and weighted to obtain a comprehensive agglomeration risk index. Based on the comprehensive agglomeration risk index, the risk level of the areas where agglomeration risk exists is determined.

6. The industrial silicon production control system based on multi-parameter collaborative optimization according to claim 5, characterized in that, A coupling identification strategy for agglomeration risk is established based on the drying temperature state factor and the drying dust state factor to screen and identify areas where agglomeration risk exists. Specifically, a multiple coupling identification strategy for agglomeration risk is established based on the drying temperature state factor and the drying dust state factor, and the areas where agglomeration risk exists are marked in turn. If a monitoring area is marked as an area where agglomeration risk exists three times, then the corresponding monitoring area is identified as an area where agglomeration risk exists.

7. The industrial silicon production control system based on multi-parameter collaborative optimization according to claim 6, characterized in that, The multi-stage block risk coupling identification strategy includes a primary block risk identification strategy, a secondary block risk identification strategy, and a tertiary block risk identification strategy. The specific strategy for identifying agglomeration risk is as follows: compare the drying temperature state factor with the preset temperature drying threshold. If the drying temperature state factor is greater than or equal to the preset temperature drying threshold, then mark the corresponding monitoring area as a non-agglomeration risk area. If the drying temperature state factor is less than the preset temperature drying threshold, the corresponding monitoring area will be marked as an area where the risk of clumping exists. The secondary agglomeration risk identification strategy is as follows: compare the first dust state factor with the preset first dust drying threshold. If the first dust state factor is greater than or equal to the preset first dust drying threshold, then mark the corresponding monitoring area as a non-agglomeration risk area. If the first dust state factor is less than the preset first dust drying threshold, the corresponding monitoring area will be marked as an area where the risk of agglomeration exists. The three-stage agglomeration risk identification strategy is as follows: the second dust state factor is compared with the preset second dust drying threshold. If the second dust state factor is greater than or equal to the preset second dust drying threshold, the corresponding monitoring area is marked as a non-agglomeration risk area. If the second dust state factor is less than the preset second dust drying threshold, the corresponding monitoring area will be marked as an area where the risk of agglomeration exists.

8. The industrial silicon production control system based on multi-parameter collaborative optimization according to claim 7, characterized in that, The drying effect assessment module evaluates the overall drying effect by comprehensively considering the drying status factors of each monitoring area. Based on the overall drying effect, it triggers collaborative intervention operations, and then performs secondary analysis on the drying status factors after the collaborative intervention operations to determine whether the drying operations are qualified. The specific details are as follows: By summarizing the drying status factors of each area with clumping risk in real time, the drying status effect coefficient is obtained, and the overall drying effect is evaluated based on the drying status effect coefficient. Specifically, the drying status effect coefficient is compared with the drying effect threshold. If the drying status effect coefficient is greater than or equal to the drying effect threshold, the drying operation is unqualified. If the drying status effect coefficient is less than the drying effect threshold, the drying operation is qualified and drying continues. When the drying effect is unqualified, a collaborative intervention operation is triggered. The drying status factors after the collaborative intervention operation are analyzed again to determine whether the drying operation is qualified. An alarm reminder is triggered for unqualified drying operations.

9. An industrial silicon production control method based on multi-parameter collaborative optimization, applied to an industrial silicon production control system based on multi-parameter collaborative optimization as described in any one of claims 1-8, characterized in that, The specific steps are as follows: By deploying a multi-source sensor network inside the drying equipment, the internal monitoring area of ​​the drying equipment is divided, and the multi-source drying distribution parameters within the monitoring area are extracted respectively. Based on the preprocessing of the multi-source drying distribution parameters, the multi-source drying distribution parameters at continuous time are dynamically sampled to form a multi-dimensional time-series state sequence, and the multi-source drying change characteristics corresponding to each monitoring area are extracted respectively. A drying status analysis model is constructed based on the multi-source drying change characteristics within the monitoring area to obtain drying status factors in real time. Based on drying status factors, areas with agglomeration risk are identified, and the risk level of these areas is determined. The overall drying effect is evaluated by comprehensively considering the drying status factors of each monitoring area. Based on the overall drying effect, a collaborative intervention operation is triggered. The drying status factors after the collaborative intervention operation are analyzed again to determine whether the drying operation is qualified.

10. A readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it is used to implement the steps of an industrial silicon production control system based on multi-parameter collaborative optimization as described in any one of claims 1-8.

Images