Method and device for purifying tail gas for hydrogen phosphide production, equipment and storage medium

Abstract

The application provides a tail gas purification control method and device, equipment and storage medium for phosphine production, and belongs to the technical field of automatic control. The method comprises the following steps: obtaining running state data of a tail gas purification system at multiple time points in a current period; the running state data comprises inlet gas concentration, outlet gas concentration and inlet-outlet pressure difference; determining the execution sequence of absorption liquid flow regulation, catalyst bed temperature regulation and gas flow rate regulation in the tail gas purification system based on the running state data at the multiple time points, and the regulation direction of the absorption liquid flow regulation, the catalyst bed temperature regulation and the gas flow rate regulation; and performing control regulation of the tail gas purification system based on the execution sequence and the regulation direction. The tail gas purification control method and device, equipment and storage medium provided by the application can improve the tail gas purification effect.

Description

Technical Field

[0001] This application belongs to the field of automatic control technology, and more specifically, relates to a method, apparatus, equipment, and storage medium for purifying and controlling tail gas in phosphine production. Background Technology

[0002] Phosphine (PH3), a gas with unique chemical properties, has wide applications in the semiconductor and electronics industry, agricultural storage, and chemical synthesis. The exhaust gas from phosphine production typically contains unreacted raw materials, residual phosphine, and byproducts (such as hydrogen, dust, and trace amounts of acidic / alkaline impurities), and must be purified by an exhaust gas purification system before being released.

[0003] The core treatment unit of the exhaust gas purification system includes a catalytic bed and an absorption scrubbing device. In the catalytic bed, phosphine is converted into non-toxic phosphoric acid mist through catalytic oxidation. In the absorption scrubbing device, such as a spray absorption tower or a packed absorption tower, the absorbent comes into countercurrent contact with the exhaust gas, capturing incompletely catalyzed phosphine and phosphoric acid mist, thus achieving end-of-pipe purification. The exhaust gas purification effect can be adjusted by regulating the absorbent flow rate, catalyst bed temperature, and exhaust gas flow rate. Furthermore, these three parameters are coupled and mutually constrained; improper adjustment of one parameter can negate the effect of another, leading to a decrease in exhaust gas treatment efficiency. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, equipment, and storage medium for purifying and controlling tail gas in phosphine production, so as to improve the tail gas purification effect.

[0005] A first aspect of this application provides a method for purifying and controlling tail gas in phosphine production, comprising:

[0006] The system acquires operational status data of the exhaust gas purification system at multiple points in time within the current period; the operational status data includes inlet gas concentration, outlet gas concentration, and inlet / outlet pressure difference.

[0007] Based on the operating status data at the multiple time points, the execution order of the absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation in the exhaust gas purification system is determined, as well as the respective regulation directions of the absorbent flow rate regulation, the catalyst bed temperature regulation, and the gas flow rate regulation.

[0008] The exhaust gas purification system is controlled and adjusted based on the execution sequence and the adjustment direction.

[0009] A second aspect of this application provides a tail gas purification and control device for phosphine production, comprising:

[0010] The data acquisition module is used to acquire the operating status data of the exhaust gas purification system at multiple time points within the current period; the operating status data includes the inlet gas concentration, the outlet gas concentration, and the inlet and outlet pressure difference;

[0011] The decision module is used to determine the execution order of the absorbent flow rate regulation, catalyst bed temperature regulation and gas flow rate regulation in the exhaust gas purification system based on the operating status data at the multiple time points, as well as the regulation direction of each of the absorbent flow rate regulation, catalyst bed temperature regulation and gas flow rate regulation.

[0012] The control and adjustment module is used to control and adjust the exhaust gas purification system based on the execution sequence and the adjustment direction.

[0013] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described tail gas purification and control method for phosphine production.

[0014] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described tail gas purification and control method for phosphine production.

[0015] The beneficial effects of the phosphine production tail gas purification and control method, apparatus, equipment, and storage medium provided in this application embodiment are as follows:

[0016] In this embodiment, by monitoring the operating status data at multiple time points and capturing the dynamic changes in the operating status data, different operating conditions of the exhaust gas purification system can be determined. For different operating conditions, the execution order of absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation, as well as the adjustment direction of each of the absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation, can be set to improve the synergistic effect of absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation, thereby improving the exhaust gas purification effect. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1A schematic flowchart of a tail gas purification and control method for phosphine production provided in an embodiment of this application;

[0019] Figure 2 A structural block diagram of a tail gas purification and control device for phosphine production provided in an embodiment of this application;

[0020] Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0021] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0022] It is understood that in the embodiments of this application, data such as user information are involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of related data must comply with relevant laws, regulations and standards.

[0023] It should be noted that the terms "first," "second," etc., used in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0025] Please refer to Figure 1 , Figure 1 This is a schematic flowchart of a tail gas purification and control method for phosphine production provided in an embodiment of this application. The tail gas purification and control method for phosphine production provided in this embodiment can be executed by an electronic device, and the method may include:

[0026] S101: Obtain the operating status data of the exhaust gas purification system at multiple time points within the current period; the operating status data includes the inlet gas concentration, outlet gas concentration, and inlet / outlet pressure difference.

[0027] In this embodiment, the main component of the phosphine production tail gas is phosphine. Therefore, the inlet gas concentration of the tail gas purification system can be the inlet phosphine gas concentration, and the outlet gas concentration of the tail gas purification system can be the outlet phosphine gas concentration. The inlet and outlet pressure difference of the tail gas purification system is the pressure difference between the outlet gas pressure and the inlet gas pressure.

[0028] S102: Based on the operating status data at multiple time points, determine the execution sequence of absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation in the exhaust gas purification system, as well as the regulation direction of each of the absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation.

[0029] In this embodiment, by monitoring the operating status data at multiple time points and capturing the dynamic changes in the operating status data, different operating conditions of the exhaust gas purification system can be determined. The execution sequence of absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation, as well as the adjustment direction of each of these three regulation methods, can be set for different operating conditions. This can improve the synergistic effect of absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation, thereby improving the exhaust gas purification effect.

[0030] S103: Control and adjust the exhaust gas purification system based on the execution sequence and adjustment direction.

[0031] In this embodiment, by adjusting the absorbent flow valve, catalyst bed temperature control device, and gas delivery adjustment mechanism based on the above execution sequence and adjustment direction, the control and adjustment of the exhaust gas purification system can be realized, ensuring that the control and adjustment are always precisely matched with the real-time operation requirements, thereby improving the exhaust gas purification effect.

[0032] In one embodiment of this application, the execution order of absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation in the exhaust gas purification system is determined based on operating status data at multiple time points, as well as the respective regulation directions of absorbent flow rate regulation, catalyst bed temperature regulation, and gas flow rate regulation, including:

[0033] If the average concentration of the inlet gas in the current period is greater than the preset first concentration threshold, the rate of change of the inlet gas concentration in the current period is greater than the preset rate of change threshold, and the average concentration of the outlet gas in the current period is greater than the preset second concentration threshold, then the step of reducing the gas flow rate is executed first, then the step of increasing the absorbent flow rate is executed, and finally the step of increasing the catalyst bed temperature is executed.

[0034] If the rate of change of the inlet gas concentration in the current period is less than or equal to the preset rate of change threshold, the average value of the outlet gas concentration in the current period is greater than the preset second concentration threshold, and the pressure difference between the inlet and outlet is greater than the preset pressure difference threshold, then the step of reducing the gas flow rate is executed first, followed by the step of increasing the catalyst bed temperature, and finally the step of increasing the absorbent flow rate is executed.

[0035] In this embodiment, the corresponding adjustment strategies for two typical operating conditions of the exhaust gas purification system are as follows:

[0036] For high-load impact conditions, under these conditions, if the average inlet gas concentration during the current period is greater than a preset first concentration threshold, it indicates a large base load of exhaust gas requiring purification; if the rate of change of the inlet gas concentration during the current period is greater than a preset rate of change threshold, it indicates a rapidly increasing exhaust gas load requiring purification, which is an impact load; and if the average outlet gas concentration during the current period is greater than a preset second concentration threshold, it indicates insufficient current purification capacity. The first concentration threshold, the rate of change threshold, and the second concentration threshold are all preset constants. Those skilled in the art can design specific values ​​for these preset constants according to actual needs. For example, the first concentration threshold could be 850 ppm, the rate of change threshold 6%, and the second concentration threshold 0.3 ppm.

[0037] Under the aforementioned high-load impact conditions, the gas flow rate can be reduced first to prolong the gas residence time in the purification system, allowing the gas more time to contact the absorbent and react with the catalyst. Then, the absorbent flow rate can be increased to enhance the adsorption capacity, thereby increasing the gas removal rate per unit time and rapidly reducing the outlet gas concentration. Conversely, increasing the absorbent flow rate first and then reducing the gas flow rate can lead to absorbent accumulation, potentially causing an increase in the inlet and outlet pressure difference. Furthermore, increasing the catalyst bed temperature can compensate for catalyst activity and accelerate the catalytic reaction rate.

[0038] For stable load and abnormal pressure difference conditions, if the rate of change of the inlet gas concentration in the current period is less than or equal to the preset rate of change threshold, it indicates that the exhaust gas load to be purified is in a stable state; if the average value of the outlet gas concentration in the current period is greater than the preset second concentration threshold, it indicates that the current purification capacity is insufficient; if the pressure difference between the inlet and outlet is greater than the preset pressure difference threshold, it indicates that the flow resistance of the gas in the purification system is large.

[0039] At this point, firstly, reducing the gas flow rate can directly reduce the impact of the gas flow on the catalyst bed and absorbent system, quickly reduce system resistance, and avoid safety risks such as pipeline rupture and phosphine leakage caused by excessive resistance. Subsequently, since the load is stable, there is no need to quickly reduce the load by adjusting the absorbent flow rate, and raising the temperature can directly improve the catalyst activity, avoiding further increase in system resistance due to excessive absorbent flow rate. Finally, after the catalytic reaction is enhanced, increasing the absorbent flow rate can further improve the purification effect.

[0040] In one embodiment of this application, the steps of first reducing the gas flow rate, then increasing the absorbent flow rate, and finally increasing the catalyst bed temperature are performed, including:

[0041] Perform parameter adjustment operations multiple times until the stopping condition is met;

[0042] Each parameter adjustment operation includes:

[0043] Obtain the latest outlet gas concentration at multiple time points, and calculate the average of the latest outlet gas concentration based on the latest outlet gas concentration at multiple time points;

[0044] Reduce the gas flow rate in the first step;

[0045] After reducing the gas flow rate for a set time, the absorbent flow rate is increased in a second step.

[0046] After increasing the absorbent flow rate for a set time, the catalyst bed temperature is increased in a third step; the first, second, and third step lengths are all positively correlated with the latest average outlet gas concentration.

[0047] The stopping conditions are: the latest average value of the outlet gas concentration is less than or equal to the preset third concentration threshold; the third concentration threshold is less than the second concentration threshold.

[0048] In this embodiment, the outlet gas concentration can be gradually brought to the target through multiple rounds of parameter adjustment. Taking a high-load impact condition as an example, each time a parameter adjustment operation is performed, the latest outlet gas concentration at multiple time points is first obtained, and the average value of the latest outlet gas concentration is calculated by arithmetic mean. Based on this, the first step length is determined according to the positive correlation between the average value of the latest outlet gas concentration and the first step length. The gas flow rate is reduced by the first step length. After reducing the gas flow rate, wait for a set time (e.g., 5 minutes), and then increase the absorbent flow rate by the second step length to improve the adsorption effect. After increasing the absorbent flow rate, wait for a set time again (e.g., 5 minutes), and then increase the catalyst bed temperature by the third step length to accelerate the catalytic reaction rate.

[0049] Specifically, the first step length, the second step length, and the third step length can be calculated using the following first formula:

[0050] ;

[0051] in, This represents the length of the first step when performing the parameter adjustment operation for the t-th time. This represents the latest average concentration of the outlet gas. This indicates the third concentration threshold. This indicates the maximum value of the preset gas flow rate change (e.g., 8%). Indicates the rate of change of gas flow velocity. This represents the gas flow rate during the t-th parameter adjustment operation; This represents the second step size during the t-th execution of the parameter adjustment operation. This indicates the maximum value of the preset rate of change of absorbent flow rate (e.g., 10%). express rate of change of absorbent flow rate This represents the absorbent flow rate during the t-th parameter adjustment operation. This represents the length of the third step during the t-th execution of the parameter adjustment operation. This indicates the maximum value of the preset catalyst bed temperature adjustment step size (e.g., 5°C).

[0052] In the first formula above, the larger the average value of the latest outlet gas concentration, the larger the length of the first step, the second step, and the third step. When the outlet gas concentration is large, a larger adjustment range can achieve rapid adjustment of the exhaust gas treatment system, and when the outlet gas concentration is small, a smaller adjustment range can avoid system fluctuations.

[0053] The first step length is determined by the gas flow rate change rate and the gas flow rate during the t-th parameter adjustment operation, which can adapt to different initial gas flow rate values ​​and ensure smooth gas flow rate adjustment. Similarly, the second step length is determined by the absorbent flow rate change rate and the absorbent flow rate during the t-th parameter adjustment operation, which can adapt to different initial absorbent flow rate values ​​and ensure smooth absorbent flow rate adjustment. The catalytic efficiency of the catalyst is directly related to the absolute value of the catalyst bed temperature. The third step length is determined based on the maximum value of the preset catalyst bed temperature adjustment step length, without using a relative increase (temperature change rate). This allows for precise control of the temperature rise, avoiding overheating deactivation caused by relative increases, while also limiting high energy consumption and temperature oscillation.

[0054] It should be noted that the first step length, the second step length, and the third step length all have corresponding maximum limits. When the first step length, the second step length, and the third step length calculated using the first formula above exceed the corresponding maximum limits, the first step length, the second step length, and the third step length will be corrected to the corresponding maximum limits to avoid over-adjustment that could lead to system risks, equipment damage, or waste of resources.

[0055] As can be seen from the above, this embodiment can achieve a steady decrease in the outlet gas concentration by gradually reducing the outlet gas concentration through multiple cycles, thus avoiding failure to meet the standard due to insufficient adjustment in a single instance.

[0056] In one embodiment of this application, the exhaust gas purification system includes a pretreatment unit, a catalytic bed, and a catalyst regeneration device;

[0057] Before controlling and regulating the exhaust gas purification system based on the execution sequence and adjustment direction, the following steps are also included:

[0058] Obtain the operating parameters of the catalytic bed;

[0059] The working state of the catalyst in the catalyst bed is determined based on the working parameters of the catalyst bed; the working state of the catalyst includes normal state, mild poisoning state, moderate poisoning state or severe poisoning state;

[0060] If the catalyst is in a slightly poisoned state, adjust the operating parameters of the pretreatment unit to intercept the poison.

[0061] If the catalyst is in a moderately poisoned state, start the catalyst regeneration unit.

[0062] In this embodiment, the tail gas from phosphine production contains toxic substances such as sulfur compounds, chlorine compounds, and heavy metals. These substances can occupy active sites through chemical adsorption or corrode the catalyst support and destroy the active structure, causing catalyst poisoning and resulting in a decrease in catalytic conversion capacity.

[0063] To avoid the aforementioned problems, this embodiment can monitor the operating parameters of the catalytic bed in real time and determine the current operating status of the catalyst based on these parameters. If the catalyst is in a normal state, the flow rate of the absorbent, the temperature of the catalyst bed, and the gas flow rate in the exhaust gas purification system are adjusted normally. If the catalyst is mildly poisoned, when adjusting the flow rate of the absorbent, the temperature of the catalyst bed, and the gas flow rate in the exhaust gas purification system, it is also necessary to adjust the operating parameters of the pretreatment unit, such as increasing the frequency of adsorbent replacement in the pretreatment unit, increasing the flow rate of the spray liquid in the pretreatment unit, or optimizing the filtration accuracy, to strengthen the interception of poisons in the exhaust gas (such as sulfur, chlorine, and heavy metals), reduce the entry of poisons into the catalytic bed, and avoid aggravating the poisoning. If the catalyst is moderately poisoned, when adjusting the flow rate of the absorbent, the temperature of the catalyst bed, and the gas flow rate in the exhaust gas purification system, it is necessary to start the catalyst regeneration device, such as through thermal regeneration or chemical regeneration, to remove the poisons adsorbed on the catalyst surface and restore its activity. If the catalyst is severely poisoned, an alarm message can be output to prompt manual replacement of the catalyst.

[0064] In one embodiment of this application, determining the operating state of the catalyst in the catalytic bed based on the operating parameters of the catalytic bed includes:

[0065] The operating parameters of the catalytic bed are feature extracted to obtain the operating characteristic data of the catalytic bed;

[0066] Based on the operating characteristic data of the catalyst bed, a binary classification model is used to identify whether the catalyst is in a normal or poisoned state.

[0067] If the catalyst is in a poisoned state, based on the operating characteristic data of the catalyst bed, a multi-classification model is used to identify whether the catalyst is in a mildly poisoned, moderately poisoned, or severely poisoned state.

[0068] In this embodiment, the operating parameters of the catalytic bed may include the inlet and outlet gas concentrations, multiple temperatures of the catalytic bed, inlet and outlet oxygen content, and the inlet and outlet pressure difference. The multiple temperatures of the catalytic bed include the temperatures at multiple measuring points at the inlet, the interior, and the outlet. Feature extraction from these operating parameters yields operational characteristic data such as actual catalytic efficiency, temperature difference attenuation rate, oxygen consumption-to-temperature ratio, poison accumulation, and bed temperature uniformity.

[0069] Among them, the actual catalytic efficiency can directly characterize the activity of the catalyst; the temperature difference decay rate can characterize the degree of decay of the temperature difference between the inlet and outlet of the catalyst bed. Since the phosphine catalysis process involves a large amount of heat release, the degree of temperature difference decay can characterize the degree of decrease in reaction intensity after catalyst poisoning; the oxygen consumption-temperature difference ratio is the ratio of oxygen consumption to the temperature difference of the catalyst bed. When the catalyst is poisoned, the catalytic reaction and oxygen consumption are not synchronized, and the oxygen consumption-temperature difference ratio will increase significantly; the cumulative amount of poison is used to characterize the total mass of toxic substances adsorbed and retained by the catalyst in the catalyst bed. The cumulative amount of poison can intuitively reflect the degree of influence of poison on catalyst activity and is a key quantitative basis for the classification of poisoning degree; the bed temperature uniformity is the maximum value of the temperature deviation of multiple temperature measuring points (e.g., 3 temperature measuring points) inside the catalyst bed. When there is local catalyst poisoning in the catalyst bed, the maximum value of the temperature deviation will increase significantly.

[0070] The calculation process for each of the above-mentioned operational characteristic data is described below:

[0071] (1) Methods for determining actual catalytic efficiency, including:

[0072] The initial catalytic efficiency is determined based on the inlet gas concentration and outlet gas concentration of the catalyst bed.

[0073] The load correction factor is determined based on the relative magnitude of the inlet gas concentration of the catalyst bed and the preset standard concentration.

[0074] The initial catalytic efficiency is corrected based on the load correction factor to obtain the actual catalytic efficiency.

[0075] For example, the actual catalytic efficiency can be calculated using the following second formula:

[0076] ;

[0077] in, Indicates the actual catalytic efficiency. This indicates the concentration of the inlet gas of the catalyst bed. This indicates the concentration of the outlet gas from the catalytic bed. This indicates the rated value of the inlet gas concentration of the catalytic bed. Indicates the initial catalytic efficiency. This represents the load correction factor.

[0078] In the second formula above, high-load operating conditions are taken into account. The effect is relatively large, even if the catalyst activity has declined, because Large base number, molecule It may still remain at a relatively high value, making the catalytic efficiency appear normal and masking the problem of insufficient actual catalyst activity. Loading correction factor By using the rated value of the imported gas concentration as a benchmark, the catalytic efficiency under high load is calibrated to the rated load scenario, which can eliminate the influence of the concentration base difference. Therefore, by correcting the initial catalytic efficiency through the load correction coefficient, the actual catalytic efficiency obtained can truly reflect the insufficient catalyst activity under high load, and avoid underestimating the degree of poisoning under high load conditions.

[0079] (2) Formula for calculating the temperature difference decay rate:

[0080] ;

[0081] in, Indicates the rate of temperature difference decay. This represents the initial temperature difference between the inlet and outlet of the catalyst bed. Specifically, the average temperature difference during the stable operation of a newly added catalyst can be used as the initial temperature difference. This indicates the current temperature difference between the inlet and outlet of the catalyst bed.

[0082] (3) Formula for calculating the oxygen consumption-temperature difference ratio:

[0083] ;

[0084] in, This indicates the oxygen consumption-temperature difference ratio. This indicates the current temperature difference between the inlet and outlet of the catalyst bed. This represents the oxygen consumption, which is the difference between the oxygen content at the inlet and outlet of the catalyst bed.

[0085] (4) Formula for calculating cumulative toxicity:

[0086] ;

[0087] in, Indicates the cumulative amount of toxic substances. Indicates the mass of the catalyst. The maximum amount of poison adsorbed per unit mass of catalyst can be obtained from the catalyst's manufacturer specifications. Indicates the actual catalytic efficiency. This indicates the catalytic efficiency of the newly added catalyst.

[0088] In the above formula, This indicates the degree of decline in the catalyst's catalytic efficiency. The greater the decline in the catalyst's catalytic efficiency, the greater the amount of poison adsorbed and the greater the accumulation of poison.

[0089] (5) Formula for calculating bed temperature uniformity:

[0090] ;

[0091] in, Indicates the uniformity of bed temperature. This indicates the maximum temperature inside the catalytic bed. This represents the minimum temperature inside the catalytic bed. This represents the average temperature inside the catalyst bed, which is the average temperature of multiple temperature measuring points inside the catalyst bed.

[0092] In the above formula, ( This characterizes the degree of temperature dispersion within the catalytic bed, and is then divided by the average temperature. It can eliminate the influence of absolute temperature, making the temperature uniformity of the bed at different temperatures more comparable.

[0093] Based on the aforementioned operational characteristic data, the catalyst's operating state can be identified. In existing technologies, a multi-classification model typically outputs the catalyst's normal, mildly poisoned, moderately poisoned, or severely poisoned state directly. In this embodiment, considering that the normal state accounts for the majority of cases, a coarse screening of normal or poisoned states is first performed using a binary classification model. When the binary classification model determines that the catalyst is in a poisoned state, a multi-classification model is then used to further subdivide the poisoned state. Since the catalyst is in a normal state in most cases, the method in this embodiment only requires subdivision of the poisoned state in a few cases, thereby reducing the computational load.

[0094] The binary classification model can be implemented using existing support vector machine models or lightweight random forests, and the multi-class classification model can be implemented using existing support vector machine models or random forests.

[0095] In one embodiment of this application, the binary classification model is a first support vector machine model, and the multi-class classification model is a second support vector machine model; the first support vector machine model is trained based on a first sample dataset, and the second support vector machine model is obtained based on a second sample dataset and an expanded sample dataset; the second sample data includes sample data of mild poisoning, sample data of moderate poisoning, and sample data of severe poisoning, and the first sample dataset includes sample data of normal state and the second sample dataset;

[0096] The methods for determining the expanded sample dataset include:

[0097] Select the support vectors corresponding to the poisoning state from all support vectors of the first support vector machine model, and use them as the first target support vectors;

[0098] Feature extraction was performed on sample data of severe poisoning to obtain multiple feature vectors;

[0099] For each first target support vector, calculate the similarity between the first target support vector and multiple feature vectors respectively. If there is a feature vector with a similarity greater than the similarity threshold, then the first target support vector is used as the second target support vector.

[0100] Add random perturbations to each second target support vector to obtain an expanded sample dataset.

[0101] In this embodiment, both the binary classification model and the multi-class classification model can be implemented using a support vector machine (SVM) model. Specifically, a first SVM model can be trained based on a first sample dataset to obtain a binary classification model, and then a second SVM model can be trained based on a second sample dataset to obtain a multi-class classification model. The first and second sample datasets share sample data representing mild, moderate, and severe poisoning states.

[0102] For example, the first sample dataset includes sample data in the normal state, mild poisoning state, moderate poisoning state, and severe poisoning state. Sample data in the normal state can be labeled "normal" with a label of 0, while sample data in the mild, moderate, and severe poisoning states can be labeled "poisoned" with a label of 1. A first support vector machine model is trained based on this labeled first sample dataset.

[0103] Similarly, the second sample dataset includes samples of mild, moderate, and severe poisoning states. Samples of mild poisoning are labeled as "mild poisoning" (label 1), samples of moderate poisoning as "moderate poisoning" (label 2), and samples of severe poisoning as "severe poisoning" (label 3). A second support vector machine model is then trained based on this labeled second sample dataset.

[0104] Furthermore, considering the limited sample data of severe poisoning, this embodiment expands the data based on the training results of the first support vector machine model to obtain an expanded sample dataset, and trains a second support vector machine model based on the second sample dataset and the expanded sample dataset.

[0105] Specifically, from all support vectors in the first support vector machine model, support vectors labeled 1 (poisoned state) are selected and denoted as the first target support vector. The first target support vector is the core of the decision boundary between the normal state and the poisoned state, with strong feature representativeness, and can accurately characterize the key feature distribution of the poisoned sample. Simultaneously, following the method described in the above embodiment, feature extraction is performed on the sample data of the severely poisoned state to obtain multiple feature vectors. For each first target support vector, the similarity between the first target support vector and multiple feature vectors is calculated. If there is a feature vector with a similarity greater than a similarity threshold, it indicates that the first target support vector has a high similarity to the severely poisoned sample data, and this first target support vector is used as the second target support vector. The similarity calculation can use existing cosine similarity or Euclidean distance calculation methods. The similarity threshold is a preset constant, which can be set by those skilled in the art according to actual needs, for example, it can be set to 0.7.

[0106] Furthermore, for each second target support vector, a small random perturbation (e.g., 5% to 10% of the feature standard deviation) is added to each feature dimension, such as actual catalytic efficiency and cumulative toxicity, to ensure that the sample retains the core features after perturbation, while increasing sample diversity. The multiple second target support vectors with added random perturbation are used as the expanded sample dataset.

[0107] It should be noted that, in order to ensure the dominant role of the original severely poisoned sample data, the number of second target support vectors needs to be limited when training the second support vector machine model. For example, the number of second target support vectors cannot exceed 30% of the total number of training samples for the second support vector machine.

[0108] In one embodiment of this application, during the training process of the first support vector machine model, in order to maximize the recall rate of abnormal samples (poisoned samples), that is, to minimize the false negative probability of abnormal samples, class weights can be added to the loss function of the first support vector machine model to impose a greater penalty on the error of "classifying an abnormal sample as normal". For example, the loss function of the first support vector machine model can take the following form:

[0109] ;

[0110] in, Represents the loss function. Represents the feature weight vector. Indicates weight, This represents the feature vector of a single sample. Represents the decision function. The condition was determined to be poisoning. This is considered a normal state. This is an L2 regularization term used to prevent the model from overfitting; Indicates the penalty coefficient. This represents the total number of samples in the first sample dataset. This represents the class weight of the i-th sample, where k=0 represents a normal sample. k=1 represents a poisoned sample, corresponding to , This represents the true label of the sample (0 or 1).

[0111] In the above formula, when the first support vector machine model misclassifies a poisoned sample as a normal sample... , If positive, then multiply by The total loss will increase sharply, prompting model optimization. and This reduces such errors; conversely, when the first support vector machine model misclassifies normal samples as poisoned samples, although it will also incur losses, The loss ratio is low, allowing for a small number of misclassifications. Therefore, by setting the above loss function, the first support vector machine model is made to prioritize correcting normal samples as poisoned samples over correcting poisoned samples during training, thus further improving the reliability of catalyst poisoning detection.

[0112] Based on the same inventive concept, this application also provides a phosphine production tail gas purification control device for implementing the above-mentioned phosphine production tail gas purification control method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the phosphine production tail gas purification control device provided below can be found in the limitations of the phosphine production tail gas purification control method described above, and will not be repeated here.

[0113] This application provides a tail gas purification and control device for phosphine production, such as... Figure 2 As shown, the tail gas purification and control device 20 for phosphine production includes: a data acquisition module 21, a decision module 22, and a control and regulation module 23.

[0114] The data acquisition module 21 is used to acquire the operating status data of the exhaust gas purification system at multiple time points within the current period; the operating status data includes the inlet gas concentration, the outlet gas concentration, and the inlet and outlet pressure difference;

[0115] The decision module 22 is used to determine the execution order of the absorbent flow rate regulation, catalyst bed temperature regulation and gas flow rate regulation in the exhaust gas purification system based on the operating status data at multiple time points, as well as the regulation direction of each of the absorbent flow rate regulation, catalyst bed temperature regulation and gas flow rate regulation.

[0116] The control and adjustment module 23 is used to control and adjust the exhaust gas purification system based on the execution sequence and adjustment direction.

[0117] In one embodiment of this application, the decision module 22 is specifically used for:

[0118] If the average concentration of the inlet gas in the current period is greater than the preset first concentration threshold, the rate of change of the inlet gas concentration in the current period is greater than the preset rate of change threshold, and the average concentration of the outlet gas in the current period is greater than the preset second concentration threshold, then the step of reducing the gas flow rate is executed first, then the step of increasing the absorbent flow rate is executed, and finally the step of increasing the catalyst bed temperature is executed.

[0119] If the rate of change of the inlet gas concentration in the current period is less than or equal to the preset rate of change threshold, the average value of the outlet gas concentration in the current period is greater than the preset second concentration threshold, and the pressure difference between the inlet and outlet is greater than the preset pressure difference threshold, then the step of reducing the gas flow rate is executed first, followed by the step of increasing the catalyst bed temperature, and finally the step of increasing the absorbent flow rate is executed.

[0120] In one embodiment of this application, the decision module 22 is further configured to:

[0121] Perform parameter adjustment operations multiple times until the stopping condition is met;

[0122] Each parameter adjustment operation includes:

[0123] Obtain the latest outlet gas concentration at multiple time points, and calculate the average of the latest outlet gas concentration based on the latest outlet gas concentration at multiple time points;

[0124] Reduce the gas flow rate in the first step;

[0125] After reducing the gas flow rate for a set time, the absorbent flow rate is increased in a second step.

[0126] After increasing the absorbent flow rate for a set time, the catalyst bed temperature is increased in a third step; the first, second, and third step lengths are all positively correlated with the latest average outlet gas concentration.

[0127] The stopping conditions are: the latest average value of the outlet gas concentration is less than or equal to the preset third concentration threshold; the third concentration threshold is less than the second concentration threshold.

[0128] In one embodiment of this application, the exhaust gas purification system includes a pretreatment unit, a catalytic bed, and a catalyst regeneration device; before the exhaust gas purification system is controlled and adjusted based on the execution sequence and adjustment direction, the control and adjustment module 23 is specifically used for:

[0129] Obtain the operating parameters of the catalytic bed;

[0130] The working state of the catalyst in the catalyst bed is determined based on the working parameters of the catalyst bed; the working state of the catalyst includes normal state, mild poisoning state, moderate poisoning state or severe poisoning state;

[0131] If the catalyst is in a slightly poisoned state, adjust the operating parameters of the pretreatment unit to intercept the poison.

[0132] If the catalyst is in a moderately poisoned state, start the catalyst regeneration unit.

[0133] In one embodiment of this application, the control adjustment module 23 is further configured to:

[0134] The operating parameters of the catalytic bed are feature extracted to obtain the operating characteristic data of the catalytic bed;

[0135] Based on the operating characteristic data of the catalyst bed, a binary classification model is used to identify whether the catalyst is in a normal or poisoned state.

[0136] If the catalyst is in a poisoned state, based on the operating characteristic data of the catalyst bed, a multi-classification model is used to identify whether the catalyst is in a mildly poisoned, moderately poisoned, or severely poisoned state.

[0137] In one embodiment of this application, the binary classification model is a first support vector machine model, and the multi-classification model is a second support vector machine model; the first support vector machine model is trained based on a first sample dataset, and the second support vector machine model is trained based on a second sample dataset and an expanded sample dataset; the second sample data includes sample data of mild poisoning, sample data of moderate poisoning, and sample data of severe poisoning; the first sample dataset includes sample data of normal state and the second sample dataset; the control and adjustment module 23 is further configured to:

[0138] Select the support vectors corresponding to the poisoning state from all support vectors of the first support vector machine model, and use them as the first target support vectors;

[0139] Feature extraction was performed on sample data of severe poisoning to obtain multiple feature vectors;

[0140] For each first target support vector, calculate the similarity between the first target support vector and multiple feature vectors respectively. If there is a feature vector with a similarity greater than the similarity threshold, then the first target support vector is used as the second target support vector.

[0141] Add random perturbations to each second target support vector to obtain an expanded sample dataset.

[0142] In one embodiment of this application, the operating parameters of the catalytic bed include the inlet gas concentration and the outlet gas concentration, and the operating characteristic data of the catalytic bed include the actual catalytic efficiency; the control and adjustment module 23 is further used for:

[0143] The initial catalytic efficiency is determined based on the inlet gas concentration and outlet gas concentration of the catalyst bed.

[0144] The load correction factor is determined based on the relative magnitude of the inlet gas concentration of the catalyst bed and the preset standard concentration.

[0145] The initial catalytic efficiency is corrected based on the load correction factor to obtain the actual catalytic efficiency.

[0146] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of each module / unit in the above-described device embodiments, for example... Figure 2 The functions of the data acquisition module 21, decision-making module 22, and control and regulation module 23 are shown.

[0147] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0148] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0149] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store preset constants such as a first concentration threshold, a rate of change threshold, and a second concentration threshold.

[0150] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the phosphine production tail gas purification control method provided in the embodiments of this application, or they can execute the implementation method of the electronic device described in the embodiments of this application, which will not be repeated here.

[0151] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0152] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., provided on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0153] Those skilled in the art will recognize that the modules / units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0154] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0155] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules, units, or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or modules / units, or it may be an electrical, mechanical, or other form of connection.

[0156] The modules / units described as separate components may or may not be physically separate. Similarly, the components shown as modules / units may or may not be physical modules / units; they may be located in one place or distributed across multiple network modules / units. Some or all of the modules / units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0157] Furthermore, the functional modules / units in the various embodiments of this application can be integrated into one processing module / unit, or each module / unit can exist physically separately, or two or more modules / units can be integrated into one module / unit. The integrated modules / units described above can be implemented in hardware or in the form of software functional modules / units.

[0158] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for controlling tail gas purification for phosphine production, characterized by, The method comprises: acquiring operation state data of the tail gas purification system at multiple time points in a current period; the operation state data comprises inlet gas concentration, outlet gas concentration and pressure difference between inlet and outlet; determining execution sequence of absorbent flow adjustment, catalyst bed temperature adjustment and gas flow rate adjustment in the tail gas purification system based on the operation state data at the multiple time points, and adjustment direction of each of the absorbent flow adjustment, the catalyst bed temperature adjustment and the gas flow rate adjustment; controlling and adjusting the tail gas purification system based on the execution sequence and the adjustment direction; wherein the determination of the execution sequence of the absorbent flow adjustment, the catalyst bed temperature adjustment and the gas flow rate adjustment in the tail gas purification system based on the operation state data at the multiple time points, and the adjustment direction of each of the absorbent flow adjustment, the catalyst bed temperature adjustment and the gas flow rate adjustment comprises: if the average value of the inlet gas concentration in the current period is greater than a preset first concentration threshold value, the change rate of the inlet gas concentration in the current period is greater than a preset change rate threshold value, and the average value of the outlet gas concentration in the current period is greater than a preset second concentration threshold value, the step of reducing the gas flow rate is performed first, then the step of increasing the absorbent flow rate is performed, and finally the step of increasing the catalyst bed temperature is performed; if the change rate of the inlet gas concentration in the current period is less than or equal to the preset change rate threshold value, the average value of the outlet gas concentration in the current period is greater than the preset second concentration threshold value, and the pressure difference between the inlet and the outlet is greater than a preset pressure difference threshold value, the step of reducing the gas flow rate is performed first, then the step of increasing the catalyst bed temperature is performed, and finally the step of increasing the absorbent flow rate is performed.

2. The tail gas purification control method for production of phosphine according to Claim 1, characterized by, The step of reducing the gas flow rate is performed first, then the step of increasing the absorbent flow rate is performed, and finally the step of increasing the catalyst bed temperature is performed, which comprises: performing the parameter adjustment operation multiple times until a stop condition is met; wherein each parameter adjustment operation comprises: acquiring the outlet gas concentration at the latest multiple time points, and calculating the average value of the latest outlet gas concentration based on the outlet gas concentration at the latest multiple time points; reducing the gas flow rate by a first step length; after a set time length of reducing the gas flow rate, increasing the absorbent flow rate by a second step length; after a set time length of increasing the absorbent flow rate, increasing the catalyst bed temperature by a third step length; the first step length, the second step length and the third step length are in a positive correlation with the average value of the latest outlet gas concentration; the stop condition is that the average value of the latest outlet gas concentration is less than or equal to a preset third concentration threshold value; the third concentration threshold value is less than the second concentration threshold value.

3. The tail gas purification control method for hydrogen phosphide production according to Claim 1, characterized by, The tail gas purification system comprises a pretreatment unit, a catalyst bed and a catalyst regeneration device; before the control and adjustment of the tail gas purification system based on the execution sequence and the adjustment direction, the method further comprises: acquiring working parameters of the catalyst bed; determining working state of the catalyst in the catalyst bed based on the working parameters of the catalyst bed; the working state of the catalyst comprises normal state, mild poisoning state, moderate poisoning state or severe poisoning state; if the working state of the catalyst is a mild poisoning state, adjusting the working parameters of the pretreatment unit to intercept the poison; if the working state of the catalyst is a moderate poisoning state, starting the catalyst regeneration device.

4. The tail gas purification control method for production of phosphine according to Claim 3, characterized by, The working state of the catalyst in the catalytic bed is determined based on the working parameters of the catalytic bed, comprising: characteristic extraction is performed on the working parameters of the catalytic bed to obtain operating characteristic data of the catalytic bed; based on the operating characteristic data of the catalytic bed, and through a binary classification model, the catalyst is identified as a normal state or a poisoning state; if the catalyst is in a poisoning state, based on the operating characteristic data of the catalytic bed, and through a multi-classification model, the catalyst is identified as a mild poisoning state, a moderate poisoning state or a severe poisoning state.

5. The tail gas purification control method for production of phosphine according to claim 4, characterized by, The binary classification model is a first support vector machine model, and the multi-classification model is a second support vector machine model; the first support vector machine model is trained based on a first sample data set, and the second support vector machine model is obtained based on a second sample data set and an expanded sample data set; the second sample data includes sample data of a mild poisoning state, sample data of a moderate poisoning state and sample data of a severe poisoning state, and the first sample data set includes sample data of a normal state and the second sample data set; The determination method of the expanded sample data set comprises: selecting the support vectors corresponding to the poisoning state from all the support vectors of the first support vector machine model as first target support vectors; characteristic extraction is performed on the sample data of the severe poisoning state to obtain a plurality of feature vectors; for each first target support vector, the similarity between the first target support vector and the plurality of feature vectors is calculated, and if there is a feature vector with a similarity greater than a similarity threshold, the first target support vector is taken as a second target support vector; a random disturbance is added to each second target support vector to obtain the expanded sample data set.

6. The tail gas purification control method for production of phosphine according to claim 4, characterized by, The working parameters of the catalytic bed include the inlet gas concentration and the outlet gas concentration of the catalytic bed, and the operating characteristic data of the catalytic bed includes the actual catalytic efficiency, and the determination method of the actual catalytic efficiency comprises: determining the initial catalytic efficiency based on the inlet gas concentration and the outlet gas concentration of the catalytic bed; determining a load correction coefficient based on the relative size of the inlet gas concentration of the catalytic bed and the preset standard concentration; correcting the initial catalytic efficiency based on the load correction coefficient to obtain the actual catalytic efficiency.

7. A tail gas purification control device for phosphine production, characterized by comprising: a phosphine production device; a tail gas purification device; and a control device. comprising: a data acquisition module for acquiring operating state data of the tail gas purification system at a plurality of time points in a current period; The operating state data includes inlet gas concentration, outlet gas concentration and inlet and outlet pressure difference; a decision module for determining the execution order of the absorbent flow adjustment, the catalyst bed temperature adjustment and the gas flow rate adjustment in the tail gas purification system based on the operating state data at the plurality of time points, and the adjustment direction of each of the absorbent flow adjustment, the catalyst bed temperature adjustment and the gas flow rate adjustment; A control adjustment module is configured to perform control adjustment of the tail gas purification system based on the execution sequence and the adjustment direction. The decision module is specifically configured to: If the average value of the inlet gas concentration in the current period is greater than a preset first concentration threshold, the rate of change of the inlet gas concentration in the current period is greater than a preset rate of change threshold, and the average value of the outlet gas concentration in the current period is greater than a preset second concentration threshold, the step of reducing the gas flow rate is performed first, then the step of increasing the absorbent flow rate is performed, and finally the step of increasing the catalyst bed temperature is performed. If the rate of change of the inlet gas concentration in the current period is less than or equal to a preset rate of change threshold, the average value of the outlet gas concentration in the current period is greater than a preset second concentration threshold, and the inlet and outlet pressure difference is greater than a preset pressure difference threshold, the step of reducing the gas flow rate is performed first, then the step of increasing the catalyst bed temperature is performed, and finally the step of increasing the absorbent flow rate is performed.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. The computer program is executed by the processor to implement the steps of the method according to any one of claims 1 to 6.

Images