Method and device for determining corrosion condition of equipment, equipment and storage medium

Abstract

The embodiment of the invention provides an equipment corrosion condition determination method and device, equipment and a storage medium. The method comprises the steps of obtaining first fluid state data used for determining the corrosion condition, inputting the first fluid state data into a corrosion condition determination model, outputting predicted corrosion index data corresponding to the first fluid state data, and determining and displaying the corrosion condition corresponding to target equipment based on the predicted corrosion index data. According to the technical scheme, the problem that the corrosion state of the oil refining device cannot be accurately judged in time in the prior art is solved, automatic monitoring of the corrosion condition is achieved, the accuracy and precision of the determined result can be guaranteed, and the safety and reliability of equipment are guaranteed.

Description

Technical Field

[0001] This disclosure relates to the field of petrochemical production technology, and in particular to a method, apparatus, equipment, and storage medium for determining the corrosion status of equipment. Background Technology

[0002] In petrochemical enterprises, regular and ad-hoc equipment maintenance is a crucial aspect of equipment management. Refining units, in particular, are highly susceptible to corrosion due to the long-term exposure to impurities such as sulfur, nitrogen, and salts, as well as harsh environmental conditions including high temperatures and pressures. This corrosion can lead to decreased mechanical performance and potentially cause leaks, explosions, and other safety incidents. Therefore, regular inspections of refining units are essential to assess their corrosion status and enable timely repairs when corrosion becomes severe.

[0003] In related technologies, the corrosion status of oil refining units is judged by indicators such as sulfur dew point and water dew point. However, these indicators can only be collected and analyzed by specialized equipment, which is quite complicated to calculate, or the range is judged based on the experience of production personnel. This results in low accuracy and insufficient reliability in judging the corrosion status of oil refining units, thus preventing the equipment from being maintained in a timely and effective manner. Summary of the Invention

[0004] This disclosure provides a method, apparatus, device, and storage medium for determining the corrosion status of equipment, in order to solve the problem in the related art that the corrosion status of oil refining equipment cannot be determined in a timely and accurate manner.

[0005] In a first aspect, embodiments of this disclosure provide a method for determining the corrosion status of equipment, the method comprising:

[0006] Acquire first fluid state data to determine the corrosion condition. The first fluid state data is used to represent the state data of the fluid in the target device.

[0007] The first fluid state data is input into the corrosion condition determination model, and the predicted corrosion index data corresponding to the first fluid state data is output. The corrosion condition determination model is a multi-layer feedforward neural network trained based on historical fluid state data. The first historical fluid state data is recorded based on historical fluid state data, and the predicted corrosion index data includes sulfur dew point value.

[0008] Based on the predicted corrosion index data, the corrosion status of the target equipment is determined and displayed.

[0009] Secondly, embodiments of this disclosure provide an apparatus for determining the corrosion status of a device, the determination of the corrosion status of the device including:

[0010] The determination module is used to acquire first fluid state data for determining the corrosion situation. The first fluid state data is used to represent the state data of the fluid in the target device.

[0011] The processing module is used to input the first fluid state data into the corrosion condition determination model and output the predicted corrosion index data corresponding to the first fluid state data. The corrosion condition determination model is a multi-layer feedforward neural network trained based on historical fluid state data. The first historical fluid state data is recorded based on historical fluid state data, and the predicted corrosion index data includes sulfur dew point value.

[0012] The analysis module is used to determine the corrosion status of the target equipment based on the predicted corrosion index data.

[0013] Thirdly, embodiments of this disclosure also provide a control device, which includes:

[0014] At least one processor;

[0015] and memory that is communicatively connected to at least one processor;

[0016] The memory stores instructions that can be executed by at least one processor to cause the control device to perform a method for determining the corrosion status of the device as described in the first aspect of this disclosure.

[0017] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the corrosion determination method for a device as described in the first aspect of this disclosure.

[0018] Fifthly, embodiments of this disclosure also provide a computer program product comprising computer execution instructions, which, when executed by a processor, are used to implement the corrosion determination method for a device as described in the first aspect of this disclosure.

[0019] The corrosion determination method, apparatus, device, and storage medium provided in this disclosure acquire first fluid state data for determining corrosion, input the first fluid state data into a corrosion determination model, output predicted corrosion index data corresponding to the first fluid state data, and then determine and display the corrosion status of the target equipment based on the predicted corrosion index data. Therefore, by collecting various fluid data from the target equipment in real time, the corrosion status of the target equipment can be automatically determined, thereby achieving automated monitoring of corrosion and ensuring the accuracy and precision of the determined results, thus guaranteeing the safety and reliability of the equipment. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0021] Figure 1 An application scenario diagram of the method for determining the corrosion status of equipment provided in the embodiments of this disclosure;

[0022] Figure 2 A flowchart illustrating a method for determining the corrosion status of a device according to an embodiment of this disclosure;

[0023] Figure 3a A flowchart of a method for determining the corrosion status of a device provided in yet another embodiment of this disclosure;

[0024] Figure 3b for Figure 3a The flowchart of the method for determining the number of nodes in the target model provided in the illustrated embodiment is shown.

[0025] Figure 3c for Figure 3a The flowchart of the method for determining the training dataset and test dataset provided in the illustrated embodiment;

[0026] Figure 3d for Figure 3a The flowchart of the optimization method for the target model provided in the embodiment shown is as follows;

[0027] Figure 4 A schematic diagram of the corrosion analysis device provided in yet another embodiment of this disclosure;

[0028] Figure 5 This is a schematic diagram of the structure of a control device provided in one embodiment of the present disclosure.

[0029] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0030] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0031] The technical solutions of this disclosure and how they solve the aforementioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this disclosure will now be described with reference to the accompanying drawings.

[0032] In petrochemical enterprises, regular and ad-hoc equipment maintenance is a crucial aspect of equipment management. Refining units, in particular, are highly susceptible to corrosion due to the long-term exposure to impurities such as sulfur, nitrogen, and salts, as well as harsh environmental conditions including high temperatures and pressures. This corrosion can lead to decreased mechanical performance and potentially cause leaks, explosions, and other safety incidents. Therefore, regular inspections of refining units are essential to assess their corrosion status and enable timely repairs when corrosion becomes severe.

[0033] In related technologies, there is no corresponding calculation formula between the corrosion state of the internal structure of an oil refining unit and the detection data that can be directly obtained during the oil refining process. Therefore, it is impossible to determine the corrosion state of the oil refining unit based on the detection data obtained during the production process. Judgment can only be based on the experience of production personnel, or, during shutdown maintenance, specialized equipment can be inserted into the oil refining unit to collect and analyze data to calculate indicators such as sulfur dew point and water dew point, and then use these indicators to determine the corrosion state of the oil refining unit. This calculation is cumbersome, resulting in low accuracy and insufficient reliability in assessing the corrosion status of the oil refining unit, thus hindering timely and effective equipment maintenance.

[0034] To address this issue, this disclosure provides a method for determining the corrosion status of equipment. By using a pre-trained corrosion status determination model and based on real-time collected fluid state data, the corrosion status of the equipment can be effectively judged, ensuring automated real-time monitoring of the corrosion status of oil refining equipment and significantly improving the safety and reliability of the equipment.

[0035] The application scenarios of the embodiments of this disclosure are explained below:

[0036] Figure 1 This diagram illustrates an application scenario of the method for determining the corrosion status of equipment provided in this embodiment of the disclosure. Figure 1 As shown, during the corrosion analysis process, the server 100 receives the first fluid state data 110 collected in real time and inputs it into the corrosion determination model 120 to obtain the corresponding predicted corrosion index data 130, so as to determine and display the corrosion status of the target equipment based on the predicted corrosion index data 130.

[0037] It should be noted that, Figure 1In the scenario shown, only one of the server, the first fluid state data, and the maintenance index data prediction model is used as an example for illustration, but this disclosure is not limited to this. That is to say, the number of servers, the first fluid state data, and the maintenance index data prediction models can be arbitrary.

[0038] The following detailed description of the method for determining the corrosion status of the equipment provided in this disclosure is illustrated through specific embodiments.

[0039] Figure 2 A flowchart illustrating a method for determining the corrosion status of a device according to an embodiment of this disclosure. Figure 2 As shown, the method for determining the corrosion status of equipment provided in this embodiment includes the following steps:

[0040] Step S201: Obtain first fluid state data for determining corrosion conditions.

[0041] The first fluid state data is used to represent the state data of the fluid within the target device.

[0042] Specifically, this embodiment describes the method and steps for determining the corrosion status of equipment.

[0043] The corrosion status of equipment mainly refers to the degree of corrosion inside target equipment such as oil refining equipment and chemical equipment where liquids flow. The fluid can be crude oil or crude oil processing products (such as gasoline and natural gas).

[0044] The corrosion status of the equipment refers to the overall corrosion status within the target equipment, or the corrosion status of a specific area between adjacent sensors within the equipment (because the fluid flows within the target equipment rather than being fixed in a certain location, the corrosion caused by the fluid is usually the overall or partial corrosion within the equipment, rather than a situation where only a certain point is corroded while other locations are not). Managers can obtain the overall corrosion status, and when the overall corrosion status is relatively severe, they can then use appropriate specialized testing equipment to conduct targeted inspections of the target equipment.

[0045] The first fluid state data is the data that can be obtained by configuring existing measurement sensors in the target device, such as the temperature, pressure, and flow rate of the fluid passing through the device, as well as the composition, hydrogen sulfide content, and water vapor content that can be obtained by configuring corresponding sensors.

[0046] To ensure the accuracy of the detection, it is generally required that at least the first fluid state data include temperature and pressure data (the more types of data included, the higher the accuracy of the analysis results).

[0047] The first fluid state data is data acquired in real time. Based on the real-time first fluid state data, the corrosion status inside the target equipment is determined, ensuring the timeliness and accuracy of monitoring the inside of the target equipment.

[0048] Depending on the location of the measuring sensors configured within the target device, each type of the first fluid state data typically exists in multiple forms (but may also exist in only one form). For example, the fluid temperature flowing into the heat exchanger, the fluid temperature at a specific location within the heat exchanger, and the fluid temperature flowing out of the heat exchanger (in which case multiple forms of temperature data exist simultaneously), or the flow rate of the fluid flowing into the storage tank (in which case only one form of flow rate data exists).

[0049] For the same target device, the types and quantities of the first fluid state data used to determine the corrosion status are fixed (because the number and location of sensors in the same target device are fixed).

[0050] Step S202: Input the first fluid state data into the corrosion condition determination model and output the predicted corrosion index data corresponding to the first fluid state data.

[0051] The corrosion determination model is a multi-layer feedforward neural network trained based on historical fluid state data. The first historical fluid state data is recorded based on the historical fluid state data, and the predicted corrosion index data includes sulfur dew point value.

[0052] Specifically, the corrosion condition determination model is used to analyze the corrosion condition within a single target device. By inputting historical fluid state data corresponding to the target device (i.e., historical fluid state data), it can identify corresponding corrosion index data based on the fluid state data. Thus, by inputting the first fluid state data into the corrosion condition determination model, it can output corresponding predicted corrosion index data.

[0053] The corrosion condition determination model is trained based on a multi-layer feedforward neural network (MLP network). This model leverages the ability of MLP networks to learn and recognize information in complex models, enabling the identification of corresponding corrosion index data based on fluid state data.

[0054] Corrosion index data include water dew point and sulfur dew point, which reflect whether the target equipment is prone to generating large amounts of moisture and acidic environment, and ammonium salt crystallization point, which reflects whether the target equipment is prone to generating corrosive substances such as ammonium salts. By measuring the values ​​of these numbers, the corrosion status inside the target equipment can be determined. For example, a higher sulfur dew point indicates that there is a greater likelihood of a large accumulation of acidic fluid inside the target equipment, leading to severe corrosion.

[0055] Since corrosion index data is monitored and recorded in historical fluid state data regularly, a corrosion condition determination model can be trained by combining these corrosion index data and other data in the historical fluid state data (the relevant content will be explained in subsequent embodiments).

[0056] Step S203: Based on the predicted corrosion index data, determine and display the corrosion status of the target equipment.

[0057] Specifically, based on the real-time collected first fluid state data, the corresponding predicted corrosion index data can be calculated to determine the real-time corrosion status of the target equipment. When the real-time corrosion status meets the preset conditions, an alarm message is generated, displayed, and pushed to the management personnel, thereby effectively prompting the management personnel to carry out timely inspection and maintenance of the target equipment, thus accurately ensuring the availability of the target equipment.

[0058] The corrosion determination method for equipment provided in this disclosure acquires first fluid state data for determining corrosion, inputs the first fluid state data into a corrosion determination model, outputs predicted corrosion index data corresponding to the first fluid state data, and then determines and displays the corrosion status of the target equipment based on the predicted corrosion index data. Therefore, by collecting various fluid data from the target equipment in real time, the corrosion status of the target equipment can be automatically determined, thereby achieving automated monitoring of corrosion and ensuring the accuracy and precision of the determined results, thus guaranteeing the safety and reliability of the equipment.

[0059] Figure 3a A flowchart illustrating a method for determining the corrosion status of a device according to an embodiment of this disclosure. Figure 3a As shown, the method for determining the corrosion status of equipment provided in this embodiment includes the following steps:

[0060] Step S301: Based on the predicted corrosion index data, determine the number of input layer nodes and the number of output layer nodes in the target model.

[0061] Specifically, this embodiment mainly describes the training process of the corrosion determination model.

[0062] In this embodiment of the disclosure, the corrosion determination model is based on the MLP model. Therefore, it is necessary to first determine the structure of the MLP model (i.e., the target model, which is used to represent the untrained model corresponding to the corrosion determination model) and then determine the corresponding sample data.

[0063] When determining the structure of an MLP model, the main focus is on determining the number of nodes in the input layer and the number of nodes in the output layer of the target model.

[0064] Furthermore, such as Figure 3bThe diagram shown is a flowchart illustrating the method for determining the number of nodes in the target model, which includes the following steps:

[0065] Step S3011: The number of types of predicted corrosion index data is used as the number of output layer nodes.

[0066] Specifically, the output layer nodes of the MLP model need to output various pre-defined predicted corrosion index data. In this case, each node corresponds to one type of predicted corrosion index data. Therefore, the number of types of predicted corrosion index data can be used as the number of output layer nodes.

[0067] In some embodiments, the corrosion condition determination model outputs only one data point for each predicted corrosion index. Therefore, the number of predicted corrosion index data points can also be used as the number of output layer nodes. For example, if only the predicted corrosion index data corresponding to the sulfur dew point needs to be output, that is, only an upper limit value and a lower limit value corresponding to the sulfur dew point are included, then only two output layer nodes are needed.

[0068] Step S3012: Based on the types of target indicators in the predicted corrosion index data, determine the types of input data corresponding to the target model.

[0069] Specifically, for the input layer of the target model, similarly, it needs to be determined based on the type of data input into the target model. The type of data input into the target model needs to be determined based on the type of corrosion index data to be predicted (i.e., the type of target index).

[0070] For example, if the target index is the upper and lower limits of water dew point, then the types of data input into the target model, in addition to the fluid's temperature, pressure, and flow rate, also need to include water vapor content (and may also include indicators such as the percentage of components). Thus, the types of data input into the target model can be determined.

[0071] Step S3013: Determine the number of input layer nodes based on the type of input data.

[0072] Specifically, since the input data is fed into the erosion determination model in a set format during both model training and application, the type of input data can be directly determined as the number of input layer nodes.

[0073] Step S302: Based on the sample data generated by the mechanism model corresponding to the target device, determine the training dataset and test dataset of the target model.

[0074] Among them, the mechanism model is used to represent the model for simulating the process flow of the target equipment.

[0075] Specifically, due to the different types and numbers of sensors configured in different target devices (such as the lack of sulfur content detection sensors, water vapor content detection sensors, etc.), and the limited number of detections of corrosion index data and the insufficient accuracy of detection results, there are differences in the comprehensiveness of the collected data. In order to ensure the amount of data for training the target model and the accuracy of determining the corrosion situation obtained from the training model, the corresponding sample data can be simulated through the mechanism model to generate the training dataset and test dataset corresponding to the target model.

[0076] In some embodiments, the input data of the mechanism model is the input state data of the fluid input into the target device, and the output data of the mechanism model is the output state data of the fluid output from the target device; the sample data includes the simulation input data input into the mechanism model and the simulation output data calculated by the target model based on the simulation input data.

[0077] Specifically, the mechanism model can be established based on the process parameters and characteristics of the target equipment.

[0078] By using process parameters, sensor data collected from inside the target equipment, and corrosion index data detected, simulation tools generate simulation data that is input into the mechanism model. The simulation output data output from the mechanism model and the simulation input data input into the mechanism model are used together as sample data to determine the training dataset and the test dataset (the sample data at this time also includes the corresponding simulation data generated by the simulation tool based on the detected corrosion index data). This can effectively ensure the amount of data for training the target model.

[0079] The establishment of the mechanism model and the generation of simulation data are existing technologies in this field. No specific implementation method is limited here. Those skilled in the art can choose any method to implement it according to actual needs, and it will not be described in detail here.

[0080] Furthermore, such as Figure 3c The diagram shown is a flowchart illustrating the method for determining the training and test datasets, which includes the following steps:

[0081] Step S3021: Based on the type of input data, select candidate training data from the corresponding sample data in the mechanism model.

[0082] Specifically, the mechanistic model usually contains all types of data that the target model may need. Therefore, for each specific target device, data of the corresponding type can be selected from the sample data as alternative training data based on the type of input data of the target model corresponding to the target device.

[0083] Step S3022: Based on the set number of groups, determine the training dataset and test dataset from the candidate training data.

[0084] Specifically, due to the different processing capabilities of the servers used for model training, the amount of training data used to train the model will also vary. Therefore, the number of sets corresponding to the data can be pre-configured (such as 100 sets, 500 sets or more), and then the training data can be randomly selected from the candidate training data (which usually contains far more data than the set number of sets, such as 5000 sets or more). Then, the training data can be divided into training datasets and test datasets according to the pre-set number and proportion of training datasets and test datasets (for example, the amount of data in the training dataset accounts for 70% of the total amount of training data, and the test dataset accounts for 30%).

[0085] Step S303: Input the training dataset into the target model, train the target model, and obtain the trained target model.

[0086] Specifically, after determining the structure of the target model and the training dataset, the data in the training dataset can be input into the target model, and the target model can be trained by combining the sample data corresponding to the corrosion index data in the training dataset to obtain the trained target model.

[0087] Step S304: Input the test dataset into the trained target model and optimize the target model based on the set accuracy target.

[0088] Specifically, after obtaining the trained target model, it is possible to combine it with the test dataset and optimize the parts of the model other than the node parameters to ensure that the error between the target model output and the actual result (corrosion index data in the sample data) is satisfied.

[0089] Furthermore, such as Figure 3d The diagram shown is a flowchart of the optimization method for the target model, which includes the following steps:

[0090] Step S3041: Input the test dataset into the trained target model and determine the deviation between the output of the trained target model and the corresponding data in the test dataset.

[0091] Specifically, before optimizing the target model after training, it is necessary to calculate the deviation between the sample data and the model output results corresponding to each group of data in the test dataset, in order to determine whether the deviation meets the corresponding set accuracy target.

[0092] Step S3042: Based on the first accuracy target, optimize the target parameters in the target model.

[0093] The target parameters include the parameters of the activation function in the target model, the number of hidden layers in the target model, the number of nodes in each hidden layer, the number of iterations of the target function, and the learning rate.

[0094] Specifically, the first precision target can be higher than the set precision target. For example, if the set precision target requires an error of less than 5%, then the first precision target can be an error of less than 3%. In this case, the optimization of the target model will not optimize the internal node parameters of the target model, but will only optimize non-node parameters such as the activation function parameters, hidden layer data, number of nodes in the hidden layer, number of iterations of the objective function, and learning rate.

[0095] Specifically, when inputting sample data from the test dataset, non-node parameters can be automatically (e.g., by increasing or decreasing the parameter values ​​in the activation function or the number of nodes in the hidden layer at a set speed) or manually (e.g., by changing the number of hidden layers) adjusted, and the optimal results for various non-node parameters can be determined. The non-node parameters corresponding to the optimal results can then be used as the corresponding parameters of the target model, thereby achieving the optimization of the target model.

[0096] Step S305: Use the optimized target model as the corrosion condition determination model.

[0097] Specifically, the target model after node parameter training and non-node parameter optimization is a target model that can meet the set accuracy requirements. At this time, the target model can be used as a model to determine the corrosion situation.

[0098] Step S306: Obtain first fluid state data for determining corrosion conditions.

[0099] Step S307: Input the first fluid state data into the corrosion condition determination model and output the predicted corrosion index data corresponding to the first fluid state data.

[0100] Specifically, steps S306 to S307 and Figure 2 The corresponding steps in the illustrated embodiments are the same and will not be repeated here.

[0101] Step S308: The predicted corrosion index data and the corresponding first fluid state data are added to the optimization dataset as training data.

[0102] Specifically, the corrosion determination model provided in this embodiment does not remain unchanged after a one-time training. Instead, it can be periodically optimized and trained based on the first fluid state data input over a certain period of time in actual application and the output predicted corrosion index data, so as to continuously improve the prediction accuracy and reliability of the model.

[0103] Step S309: Input the optimized dataset into the corrosion condition determination model, and optimize and train the corrosion condition determination model based on the output results of the corrosion condition determination model.

[0104] Specifically, by optimizing the dataset and training the model regularly, the model can continuously improve its prediction accuracy in practical applications, enabling it to grow and effectively guarantee its reliability.

[0105] Step S310: If the predicted corrosion index data is within the first value range of the corresponding type of index data, the corrosion situation of the target equipment is determined to be the first type.

[0106] The first type of case is used to indicate that the corrosion level of the target equipment has not reached the point where maintenance is required.

[0107] Specifically, depending on the type of predicted corrosion index data output by the model and the target equipment, corresponding judgment conditions, namely the first numerical range, can be pre-configured. If the range is met (such as the sulfur dew point being lower than a certain value), the target equipment can be considered relatively safe and the corrosion is not serious.

[0108] Step S311: If the predicted corrosion index data is not within the first value range of the corresponding type of index data, determine that the corrosion situation of the target equipment is the second type.

[0109] The second type of case is used to indicate that the corrosion level of the target equipment has reached a point where maintenance is required.

[0110] Specifically, in another case, if the predicted corrosion index data is not within the first value range (such as the ammonium salt crystallization point being higher than a certain value), it can be considered that the corrosion of the target equipment is relatively serious and measures need to be taken to deal with it.

[0111] Step S312: If the corrosion situation is the second type, generate and issue the corresponding prompt information for the corrosion situation.

[0112] Specifically, in cases of severe corrosion, corresponding alarm messages (i.e., alert messages) need to be automatically generated so that management personnel can promptly arrange maintenance and anti-corrosion treatment for the target equipment. Since the predicted corrosion index data can be generated based on real-time detected primary fluid state data, alarm messages can also be generated in real time when the corrosion of the target equipment is detected to be worsening, thus ensuring the accuracy and timeliness of target equipment detection.

[0113] The corrosion determination method for equipment provided in this disclosure involves determining the structure of a corrosion determination model corresponding to the target equipment, generating training and validation datasets based on a mechanistic model, training and optimizing the model to obtain a corrosion determination model, and then generating corresponding predicted corrosion index data based on real-time detected first fluid state data to determine the degree of corrosion of the target equipment. In cases of severe corrosion, corresponding warning information is generated. Therefore, the accuracy and reliability of the prediction results based on the corrosion determination model are guaranteed, and the model is applicable to the corresponding target equipment. Furthermore, through regular optimization and training, the model's prediction accuracy and reliability are continuously improved, thereby continuously enhancing and ensuring the safety and availability of the equipment.

[0114] Figure 4 This is a schematic diagram of the structure of a device for determining the corrosion status of an apparatus provided in one embodiment of this disclosure. Figure 4 As shown, the corrosion condition determination device 400 includes: a determination module 410, a processing module 420, and an analysis module 430. Wherein:

[0115] The determination module 410 is used to acquire first fluid state data for determining the corrosion situation. The first fluid state data is used to represent the state data of the fluid in the target device.

[0116] Processing module 420 is used to input the first fluid state data into the corrosion condition determination model and output the predicted corrosion index data corresponding to the first fluid state data. The corrosion condition determination model is a multi-layer feedforward neural network trained based on historical fluid state data. The first historical fluid state data is recorded based on historical fluid state data. The predicted corrosion index data includes sulfur dew point value.

[0117] Analysis module 430 is used to determine the corrosion status of the target equipment based on the predicted corrosion index data.

[0118] Optionally, the analysis module 430 is specifically used to: if the predicted corrosion index data includes the upper and lower limits of the target index, and the target index includes at least one of sulfur dew point value, water dew point value, and ammonium salt crystal point value; if the predicted corrosion index data is within the first numerical range of the corresponding type of index data, determine the corrosion situation of the target equipment as a first type of situation, wherein the first type of situation indicates that the corrosion degree of the target equipment has not reached the level requiring maintenance; if the predicted corrosion index data is not within the first numerical range of the corresponding type of index data, determine the corrosion situation of the target equipment as a second type of situation, wherein the second type of situation indicates that the corrosion degree of the target equipment has reached the level requiring maintenance; if the corrosion situation is the second type of situation, generate and issue a prompt message corresponding to the corrosion situation.

[0119] Optionally, the processing module 420 is specifically used to obtain the corrosion condition determination model in the following ways: based on the predicted corrosion index data, determine the number of input layer nodes and the number of output layer nodes in the target model; based on the sample data generated by the mechanism model corresponding to the target equipment, determine the training dataset and test dataset of the target model, wherein the mechanism model is used to represent the model for simulating the process flow of the target equipment; input the training dataset into the target model to train the target model and obtain the trained target model; input the test dataset into the trained target model and optimize the target model based on the set accuracy target; and use the optimized target model as the corrosion condition determination model.

[0120] Optionally, the processing module 420 is specifically used to: use the number of types of predicted corrosion index data as the number of output layer nodes; determine the type of input data corresponding to the target model based on the type of target index in the predicted corrosion index data; and determine the number of input layer nodes based on the type of input data.

[0121] Optionally, the processing module 420 specifically includes: the input data of the mechanism model is the input state data of the fluid input into the target device; the output data of the mechanism model is the output state data of the fluid output from the target device; the sample data includes the simulation input data input into the mechanism model and the simulation output data calculated by the target model based on the simulation input data.

[0122] Optionally, the processing module 420 is specifically used to determine candidate training data from the corresponding sample data in the mechanism model based on the type of input data; and to determine the training dataset and test dataset from the candidate training data based on a set number of groups.

[0123] Optionally, the processing module 420 is specifically used to input the test dataset into the trained target model, determine the deviation between the output of the trained target model and the corresponding data in the test dataset; and optimize the target parameters in the target model based on the first precision target, wherein the target parameters include the parameters of the activation function in the target model, the number of hidden layers in the target model, the number of nodes in each hidden layer, the number of iterations of the target function, and the learning rate.

[0124] Optionally, the processing module 420 is further configured to: input the first fluid state data into the corrosion condition determination model, output the predicted corrosion index data corresponding to the first fluid state data, add the predicted corrosion index data and the corresponding first fluid state data as training data to the optimization dataset; input the optimization dataset into the corrosion condition determination model, and optimize and train the corrosion condition determination model based on the output results of the corrosion condition determination model.

[0125] In this embodiment, the corrosion condition determination device solves the problem of timely and accurate judgment of the corrosion status of oil refining equipment in related technologies by combining various modules, realizes automated monitoring of corrosion condition, and can ensure the accuracy and precision of the determination results, so as to ensure the safety and reliability of the equipment.

[0126] Figure 5 This is a schematic diagram of the structure of a control device provided in one embodiment of the present disclosure, as shown below. Figure 5 As shown, the control device 500 includes a memory 510 and a processor 520.

[0127] The memory 510 stores a computer program that can be executed by at least one processor 520. This computer program is executed by at least one processor 520 to enable the control device to implement the material removal method provided in any of the above embodiments or the corrosion determination method for the device provided in any of the above embodiments.

[0128] The memory 510 and the processor 520 can be connected via a bus 530.

[0129] The relevant explanations can be understood by referring to the corresponding descriptions and effects in the method embodiments, and will not be repeated here.

[0130] The relevant explanations can be understood by referring to the corresponding descriptions and effects in the method embodiments, and will not be repeated here.

[0131] One embodiment of this disclosure provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the material removal method as provided in any of the above method embodiments or the corrosion determination method of the device as provided in any of the above embodiments.

[0132] The computer-readable storage medium can be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0133] One embodiment of this disclosure provides a computer program product comprising computer-executable instructions that, when executed by a processor, are used to implement a material removal method as described in the above method embodiments or a corrosion determination method for a device as provided in any of the above embodiments.

[0134] In the several embodiments provided in this disclosure, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0135] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0136] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for determining the corrosion status of equipment, characterized in that, include: Acquire first fluid state data for determining corrosion conditions, wherein the first fluid state data represents the state data of the fluid within the target device; The first fluid state data is input into the corrosion condition determination model, and the predicted corrosion index data corresponding to the first fluid state data is output. The corrosion condition determination model is a multi-layer feedforward neural network trained based on historical fluid state data. The first historical fluid state data is recorded based on historical fluid state data. The predicted corrosion index data includes sulfur dew point value. Based on the predicted corrosion index data, the corrosion status of the target equipment is determined and displayed.

2. The method according to claim 1, characterized in that, The predicted corrosion index data includes the upper limit and lower limit of the target index, and the target index includes at least one of sulfur dew point value, water dew point value, and ammonium salt crystal point value. The step of determining the corrosion status of the target equipment based on the predicted corrosion index data includes: If the predicted corrosion index data is within the first value range of the corresponding type of index data, the corrosion situation of the target equipment is determined to be the first type of situation, wherein the first type of situation is used to indicate that the corrosion degree of the target equipment has not reached the level that requires maintenance; If the predicted corrosion index data is not within the first value range of the corresponding type of index data, the corrosion situation of the target equipment is determined to be the second type of situation, wherein the second type of situation is used to indicate that the corrosion degree of the target equipment has reached the level that requires maintenance; If the corrosion condition is classified as the second type, a corresponding prompt message will be generated and issued.

3. The method according to claim 1, characterized in that, The corrosion condition determination model was obtained in the following manner: Based on the predicted corrosion index data, determine the number of input layer nodes and output layer nodes in the target model; Based on the sample data generated by the mechanism model corresponding to the target equipment, the training dataset and test dataset of the target model are determined, wherein the mechanism model is used to represent the model for simulating the process flow of the target equipment; The training dataset is input into the target model, and the target model is trained to obtain the trained target model. The test dataset is input into the trained target model, and the target model is optimized based on the set accuracy target. The optimized target model is used as the model for determining the corrosion situation.

4. The method according to claim 3, characterized in that, The process of determining the number of input layer nodes and output layer nodes in the target model based on predicted corrosion index data includes: The number of types of predicted corrosion index data is used as the number of output layer nodes; Based on the types of target indicators in the predicted corrosion index data, the types of input data corresponding to the target model are determined. The number of input layer nodes is determined based on the type of input data.

5. The method according to claim 4, characterized in that, The input data of the mechanism model is the input state data of the fluid input into the target device, and the output data of the mechanism model is the output state data of the fluid output from the target device. The sample data includes simulation input data input into the mechanism model and simulation output data calculated by the target model based on the simulation input data.

6. The method according to claim 5, characterized in that, The sample data generated based on the target device's corresponding mechanism model determines the training dataset and test dataset of the target model, including: Based on the input data type, candidate training data are determined from the corresponding sample data in the mechanism model; Based on a set number of groups, the training dataset and the test dataset are determined from the candidate training data.

7. The method according to claim 3, characterized in that, The step of inputting the test dataset into the trained target model and optimizing the target model based on a set accuracy target includes: The test dataset is input into the trained target model, and the deviation between the output of the trained target model and the corresponding data in the test dataset is determined. Based on the first accuracy target, the target parameters in the target model are optimized, wherein the target parameters include the parameters of the activation function in the target model, the number of hidden layers in the target model, the number of nodes in each hidden layer, the number of iterations of the target function, and the learning rate.

8. The method according to any one of claims 3 to 7, characterized in that, After inputting the first fluid state data into the corrosion condition determination model and outputting the predicted corrosion index data corresponding to the first fluid state data, the method further includes: The predicted corrosion index data and the corresponding first fluid state data are used as training data and added to the optimization dataset; The optimized dataset is input into the corrosion condition determination model, and the corrosion condition determination model is optimized and trained based on the output results of the corrosion condition determination model.

9. A device for determining the corrosion status of equipment, characterized in that, The device for determining the corrosion status of the equipment includes: The determination module is used to acquire first fluid state data for determining the corrosion situation, wherein the first fluid state data is used to represent the state data of the fluid in the target device. The processing module is used to input the first fluid state data into the corrosion condition determination model and output the predicted corrosion index data corresponding to the first fluid state data. The corrosion condition determination model is a multi-layer feedforward neural network trained based on historical fluid state data. The first historical fluid state data is recorded based on historical fluid state data. The predicted corrosion index data includes sulfur dew point value. The analysis module is used to determine the corrosion status of the target equipment based on the predicted corrosion index data.

10. A control device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, cause the control device to perform the corrosion determination method for the device as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method for determining the corrosion status of the device as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, The computer program product includes computer execution instructions, which, when executed by a processor, are used to implement the method for determining the corrosion status of the device as described in any one of claims 1 to 8.

Images