Multi-parameter water bath kettle corrosion prevention monitoring method and system, water bath kettle and storage medium

Abstract

The invention discloses a multi-parameter water bath kettle corrosion prevention monitoring method and system, a water bath kettle and a storage medium, and belongs to the technical field of water bath kettles. The method comprises the following steps: acquiring material data of a water bath kettle and index data of a target medium in the water bath kettle, wherein the material data comprises a material type, a material thickness and a corrosion rate jump value of the material type under the contact of a corrosion medium; determining corrosion characteristic data of the target medium according to the material data and the index data; a corrosion prediction result is determined according to the corrosion characteristic data, the index data and a corrosion monitoring identification model, the corrosion monitoring identification model is provided with an attention layer, and a data weight corresponding to the index data is determined by the attention layer and the corrosion characteristic data; and determining a monitoring result according to the material thickness and the corrosion prediction result. The method has the beneficial effect of reducing the probability of problems caused by corrosion of the water bath kettle.

Description

Technical Field

[0001] The present invention relates to the technical field of water bath pots, and in particular to a multi-parameter water bath pot anti-corrosion monitoring method, system, water bath pot and storage medium. Background Art

[0002] Water baths are important constant-temperature devices, often used in experiments such as sample incubation, enzyme reactions, and microbial culture. However, the interior of a water bath is exposed to solutions containing various chemical substances for long periods of time, and during this process, various components are prone to corrosion. Corrosion in water baths can lead to various problems, such as corrosion of load-bearing or stress-bearing components such as brackets and pot body connections, which can weaken their strength and potentially cause sudden damage or collapse of the equipment during movement or operation, resulting in burns. Furthermore, corrosion products often lead to sample contamination. Currently, manual judgment is still used to determine the corrosion status of water baths, which increases the probability of corrosion-related problems occurring in the aforementioned water baths. Summary of the Invention

[0003] The main purpose of the present invention is to provide a multi-parameter water bath corrosion monitoring method, system, water bath, and storage medium, aiming to reduce the probability of problems caused by corrosion in water baths.

[0004] To achieve the above object, the present invention provides a multi-parameter water bath anti-corrosion monitoring method, which comprises the following steps: Acquiring material data of a water bath and index data of a target medium in the water bath, wherein the material data includes: material type, material thickness, and a corrosion rate jump value of the material type when exposed to a corrosive medium; Determining corrosion characteristic data of the target medium according to the material data and the index data; Determining a corrosion prediction result based on the corrosion characteristic data, the indicator data, and a corrosion monitoring and identification model, wherein the corrosion monitoring and identification model is provided with an attention layer, and the data weight corresponding to the indicator data is determined by the attention layer and the corrosion characteristic data; A monitoring result is determined according to the material thickness and the corrosion prediction result.

[0005] Optionally, the corrosion rate jump value includes: a strong acid corrosion jump value, a strong alkali corrosion jump value, a chloride ion corrosion jump value, and a corresponding critical temperature pitting corrosion value. The step of determining the corrosion characteristic data of the target medium according to the material data and the indicator data includes: Normalizing the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, the critical temperature pitting corrosion value, and the indicator data; Calculate the difference between the indicator data and the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, and the critical temperature pitting corrosion value to obtain a plurality of corrosion change differences; The corrosion characteristic data is determined according to a plurality of the corrosion change differences.

[0006] Optionally, the indicator data includes pH value data, chloride ion concentration data, and temperature data, and the step of calculating differences between the indicator data and the strong acid corrosion jump value, the strong base corrosion jump value, the chloride ion corrosion jump value, and the critical temperature pitting corrosion value to obtain multiple corrosion change differences includes: Calculating a first corrosion change difference according to the pH value data and the strong acid corrosion jump value; Calculating a second corrosion change difference according to the pH value data and the strong alkali corrosion jump value; Calculating a third corrosion change difference according to the chloride ion concentration data and the chloride ion corrosion jump value; A fourth corrosion variation difference is calculated based on the temperature data and the critical temperature pitting value.

[0007] Optionally, before the step of determining the corrosion prediction result based on the corrosion characteristic data, the index data and the corrosion monitoring identification model, the method further includes: Obtaining historical indicator data of the medium in the water bath before the current moment, and measuring corresponding corrosion results; Calculating historical corrosion characteristic data based on the historical indicator data and the material data; Determine and generate corresponding training sets and verification sets based on the corrosion results, historical indicator data, and historical corrosion characteristic data; Training is performed according to the training set and the preset model, and the trained model is verified according to the verification set to obtain the corrosion monitoring and recognition model.

[0008] Optionally, before the step of training according to the training set and the preset model, the step further includes: Constructing a model structure corresponding to the preset model; The model structure includes: an attention layer, which determines the feature weight corresponding to each time in the historical indicator data; The data corresponding to each time period in the historical indicator data is respectively used as the first indicator data in chronological order; The feature weight is determined based on historical corrosion feature data corresponding to historical indicator data that is temporally adjacent to the first indicator data.

[0009] Optionally, the corrosion prediction result includes corrosion thickness, and the step of determining the monitoring result according to the material thickness and the corrosion prediction result includes: The real-time material thickness of the material of the water bath is calculated according to the material thickness and the corrosion thickness.

[0010] Optionally, the step of obtaining material data of the water bath and index data of the target medium in the water bath includes: Determining the material data according to the model of the water bath; The sensor is controlled to detect the composition of the target medium in the water bath at a preset frequency to obtain the index data.

[0011] In addition, to achieve the above-mentioned purpose, the present invention also provides a water bath anti-corrosion monitoring system, the water bath anti-corrosion monitoring system comprising: an acquisition module, configured to acquire material data of a water bath and index data of a target medium in the water bath, wherein the material data includes: material type, material thickness, and a corrosion rate jump value of the material type in contact with a corrosive medium; a calculation module, configured to determine corrosion characteristic data of the target medium based on the material data and the index data; an identification module, configured to determine a corrosion prediction result based on the corrosion characteristic data, the indicator data, and a corrosion monitoring identification model, wherein the corrosion monitoring identification model is provided with an attention layer, and a data weight corresponding to the indicator data is determined by the attention layer and the corrosion characteristic data; A monitoring module is used to determine a monitoring result based on the material thickness and the corrosion prediction result.

[0012] In addition, to achieve the above-mentioned purpose, the present invention also provides a water bath, which includes: a memory, a processor, and a water bath anti-corrosion monitoring program stored in the memory and runnable on the processor, wherein the water bath anti-corrosion monitoring program is configured to implement the steps of the water bath anti-corrosion monitoring method described in any one of the above items.

[0013] In addition, to achieve the above-mentioned purpose, the present invention also provides a storage medium, on which a water bath anti-corrosion monitoring program is stored. When the water bath anti-corrosion monitoring program is executed by the processor, the steps of the water bath anti-corrosion monitoring method described in any one of the above-mentioned items are implemented.

[0014] The present invention proposes a multi-parameter water bath corrosion monitoring method. The method obtains material data of the water bath and index data of the target medium in the water bath, and determines the corrosion characteristic data of the target medium based on the material data and the index data. Compared with the need for manual observation to identify the corrosion condition of the water bath, the method can accurately identify the corrosion condition, and determine the corrosion prediction result through the corrosion characteristic data, index data and corrosion monitoring identification model. The corrosion monitoring identification model is provided with an attention layer, and the data weight corresponding to the index data is determined by the attention layer and the corrosion characteristic data. The monitoring result is determined according to the material thickness and the corrosion prediction result, thereby reducing the probability of various problems caused by water bath corrosion. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the structure of a water bath in the hardware operating environment involved in an embodiment of the present invention; Figure 2 This is a schematic flow chart of a first embodiment of the multi-parameter water bath anti-corrosion monitoring method of the present invention; Figure 3 This is a schematic flow chart of a second embodiment of the multi-parameter water bath anti-corrosion monitoring method of the present invention; Figure 4 1 is a flow chart of a fourth embodiment of the multi-parameter water bath anti-corrosion monitoring method of the present invention.

[0016] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0017] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] Reference Figure 1 , Figure 1 This is a schematic diagram of the water bath structure of the hardware operating environment involved in the embodiment of the present invention.

[0019] like Figure 1As shown, the water bath may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, an interactive device 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to facilitate communication between these components. The interactive device 1003 may include a display and an input unit, such as a keyboard. Optionally, the interactive device 1003 may also be connected to the communication bus via a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a wireless fidelity (WI-FI) interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also be a storage device independent of the processor 1001.

[0020] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the water bath, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0021] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a data storage module, a network communication module, a user interface module and a multi-parameter water bath anti-corrosion monitoring program.

[0022] exist Figure 1 In the water bath shown, the network interface 1004 is mainly used for data communication with other devices; the interactive device 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the water bath of the present invention can be set in the water bath, and the water bath calls the multi-parameter water bath anti-corrosion monitoring program stored in the memory 1005 through the processor 1001, and executes the multi-parameter water bath anti-corrosion monitoring method provided by the embodiment of the present invention.

[0023] The embodiment of the present invention provides a multi-parameter water bath anti-corrosion monitoring method, referring to Figure 2 , Figure 2 This is a flow chart of a first embodiment of a multi-parameter water bath anti-corrosion monitoring method of the present invention.

[0024] In this embodiment, the multi-parameter water bath anti-corrosion monitoring method includes: Step S1, obtaining material data of a water bath and index data of a target medium in the water bath, wherein the material data includes: material type, material thickness, and a corrosion rate jump value of the material type when exposed to a corrosive medium; In this embodiment, the water bath includes an inner liner, an outer shell, a heating element, a sensor, a seal, and the like. Preferably, the material data herein refers to the inner liner material, i.e., the material that directly contacts the target medium of the water bath. The material data includes the material type, and optionally, the material type is 304 stainless steel. Material thickness refers to the thickness of the inner liner material. The corrosion rate jump value varies for different materials and environments. For materials such as stainless steel, aluminum, and titanium, the surface oxide film primarily isolates the corrosive medium. This oxide film can exist stably within a specific range, so the corrosion effect is not significant. However, when the material is in an environment where the oxide film cannot stabilize, rapid dissolution occurs. In this embodiment, the metal oxide film of 304 stainless steel can exist stably in an environment with a pH between 1.5 and 10.5. When the pH is outside of 1.5 to 10.5, the oxide film fails, and therefore the corrosion rate variation increases significantly, generally increasing by 100 to 1000 times. The type of indicator data is not limited. Optionally, the indicator data here includes at least one of pH value and ion concentration data. Optionally, the indicator data here can be obtained by sensor detection and can be determined according to the type of experiment. For example, different experimental types require different ion concentrations in the configured solution, so the ion concentration can be determined according to different configurations.

[0025] Step S2, determining corrosion characteristic data of the target medium according to the material data and the index data; In this embodiment, the failure of the oxide film in the current environment is determined by comparing the corrosion rate jump value in the material data with the index data, thereby determining the accuracy of the corrosion characteristic data. It should be noted that the current corrosion rate jump value is generally determined through experiments under standard conditions. Common default environmental conditions can be a single type of corrosive medium under standard atmospheric pressure, room temperature of 25 degrees Celsius, etc. For complex target media, the corrosion efficiency is often not directly detectable by experimental testing. Therefore, determining the corrosion characteristic data of the target medium based on the material data and the index data is actually judging the corrosion condition of the current target medium on the water bath through data from multiple dimensions.

[0026] Step S3, determining a corrosion prediction result based on the corrosion characteristic data, the indicator data, and a corrosion monitoring and identification model, wherein the corrosion monitoring and identification model is provided with an attention layer, and the data weight corresponding to the indicator data is determined by the attention layer and the corrosion characteristic data; In this embodiment, the corrosion signature data and indicator data are used to generate corresponding input vectors, which are then input into the corrosion monitoring and identification model to obtain corrosion prediction results. In this embodiment, the purpose of setting an attention layer is to assign different data weights to high and low corrosion rates. This effectively reduces the impact of corrosion rate jumps on model predictions when the value is not reached.

[0027] Step S4: determining a monitoring result according to the material thickness and the corrosion prediction result.

[0028] In this embodiment, the current monitoring result of the water bath is determined according to the material thickness and the corrosion prediction result.

[0029] In this embodiment, by obtaining the material data of the water bath and the index data of the target medium in the water bath, and determining the corrosion characteristic data of the target medium based on the material data and the index data, compared with the need for manual observation to identify the corrosion condition of the water bath, the corrosion condition can be accurately identified, and the corrosion prediction result is determined by the corrosion characteristic data, the index data and the corrosion monitoring and identification model. The corrosion monitoring and identification model is provided with an attention layer, and the data weight corresponding to the index data is determined by the attention layer and the corrosion characteristic data. The monitoring result is determined based on the material thickness and the corrosion prediction result, thereby reducing the probability of various problems caused by water bath corrosion.

[0030] Further, based on the first embodiment, a second embodiment of the multi-parameter water bath anti-corrosion monitoring method of the present invention is proposed. In this embodiment, referring to Figure 3 The corrosion rate jump value includes: a strong acid corrosion jump value, a strong alkali corrosion jump value, a chloride ion corrosion jump value and a corresponding critical temperature pitting corrosion value. The step of determining the corrosion characteristic data of the target medium according to the material data and the indicator data includes: Step S21, normalizing the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, the critical temperature pitting corrosion value, and the indicator data; It should be noted that the corrosion rate jump value here may also include: a fluorine ion corrosion jump value, a sulfur ion corrosion jump value, etc. In this embodiment, the normalization method is not limited. Common methods that can be used include relative deviation normalization and logarithmic normalization. Different corrosion rate jump values ​​can correspond to different normalized data.

[0031] Step S22, calculating differences between the indicator data and the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, and the critical temperature pitting corrosion value, respectively, to obtain a plurality of corrosion change differences; Preferably, in this embodiment, the indicator data are normalized with the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, and the critical temperature pitting value as reference values. In the normalized data, each corrosion rate jump value is normalized to 0, and the value of the indicator data is spread around 0, so that the relative deviation between the indicator data and the important value can be intuitively reflected. The corrosion characteristic data here is used to describe the relationship between the current indicator data and the corrosion jump value as a whole, wherein, when the corrosion change difference is negative, it means that the corrosion jump value has not been reached and is in a lower corrosion rate interval. When the corrosion change difference is positive, it means that the corrosion jump value has been reached and is in a higher corrosion rate interval.

[0032] Step S23: determining the corrosion characteristic data according to the plurality of corrosion change differences.

[0033] In this embodiment, the corrosion characteristic data is obtained by calculating a plurality of corrosion change differences and corresponding weights. Preferably, the data range of the corrosion characteristic data is mapped to a non-negative interval, for example: [0, 10].

[0034] Of course, in other embodiments, a corrosion feature vector is generated as the corrosion feature data based on a plurality of corrosion change differences in a preset order.

[0035] In this embodiment, by normalizing the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, the critical temperature pitting value and the index data; calculating the difference between the index data and the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, and the critical temperature pitting value, a plurality of corrosion change differences are obtained, and the corrosion characteristic data is determined based on the plurality of corrosion change differences, thereby improving the accuracy of identifying the corrosion of the target medium.

[0036] Furthermore, based on the first embodiment or the second embodiment, a third embodiment of the multi-parameter water bath anti-corrosion monitoring method of the present invention is proposed. In this embodiment, the indicator data includes: pH value data, chloride ion concentration data, and temperature data. The step of calculating the difference between the indicator data and the strong acid corrosion jump value, the strong base corrosion jump value, the chloride ion corrosion jump value, and the critical temperature pitting value to obtain multiple corrosion change differences includes: Calculating a first corrosion change difference according to the pH value data and the strong acid corrosion jump value; Calculating a second corrosion change difference according to the pH value data and the strong alkali corrosion jump value; Calculating a third corrosion change difference according to the chloride ion concentration data and the chloride ion corrosion jump value; A fourth corrosion variation difference is calculated based on the temperature data and the critical temperature pitting value.

[0037] Specifically, the corrosion change difference may include: a first corrosion change difference, a second corrosion change difference, a third corrosion change difference, and a fourth corrosion change difference. The indicator data is data continuously detected by each sensor, and before calculating the corrosion change difference, the indicator data needs to be divided into time intervals.

[0038] Specifically, the indicator data corresponding to each moment includes pH value data, chloride ion concentration data, and temperature data. Based on two temporally adjacent indicator data, the rate of change between the corresponding data types in the indicator data is calculated to obtain a change rate. The average change rate of all change rates is calculated. When the average change rate is less than a preset change rate, the two adjacent indicator data are grouped together. Thus, in this embodiment, indicator data for similar target media can be grouped together, where the time period corresponding to the same group of data is continuous.

[0039] In other embodiments, the indicator data corresponding to each moment is generated as position coordinates in the indicator data vector space. Each position in the indicator data vector space is divided into multiple clusters using a clustering algorithm. The data corresponding to the position coordinates in each cluster is then counted. The number of position coordinates reflects the duration of the indicator data for the cluster corresponding to the target medium in the water bath. Within the same group of data, the pH value data is specifically average pH value data, the chloride ion concentration data is average chloride ion data, and the temperature data is average temperature data, i.e., the average value of each data item within the same group of data.

[0040] In this embodiment, the corrosion change difference corresponding to each time period is calculated in a grouping manner, thereby improving the accuracy of the corrosion characteristic data obtained subsequently.

[0041] Further, based on any of the above embodiments, a fourth embodiment of the multi-parameter water bath anti-corrosion monitoring method of the present invention is proposed. In this embodiment, referring to Figure 4 Before the step of determining the corrosion prediction result based on the corrosion characteristic data, index data and corrosion monitoring identification model, the method further includes: Step S301, obtaining historical indicator data of the medium in the water bath before the current moment, and measuring the corresponding corrosion results; The data type of the historical indicator data is the same as the data type of the indicator data, and a corresponding corrosion result needs to be obtained after each historical indicator data is obtained. Specifically, the method of measuring the corresponding corrosion result is not limited in this embodiment by removing the medium in the water bath and detecting the change in the thickness of the water bath.

[0042] Step S302, calculating historical corrosion characteristic data based on the historical indicator data and the material data; In this embodiment, the need for calculating the historical corrosion characteristic data is the same as the method for calculating the corrosion characteristic data in the third embodiment, thereby ensuring that the historical corrosion characteristic data is of the same type as the corrosion characteristic data.

[0043] Step S303, determining and generating corresponding training sets and verification sets based on the corrosion results, historical indicator data, and historical corrosion characteristic data; Specifically, the corrosion results, historical indicator data and historical corrosion characteristic data are taken as a data set, each set of sample data in the data set includes one corrosion result, historical indicator data within a period of time and corresponding historical corrosion characteristic data, and the data set is divided into a training set and a verification set according to a preset ratio, and both the training set and the verification set include multiple sets of sample data.

[0044] Step S304 , training is performed according to the training set and the preset model, and the trained model is verified according to the verification set to obtain the corrosion monitoring and recognition model.

[0045] In this embodiment, the preset model is trained according to the training set, and the trained model is verified according to the verification set to obtain the corrosion monitoring and recognition model, thereby improving the accuracy of model monitoring and recognition.

[0046] Furthermore, before the step of training according to the training set and the preset model, the step further includes: Constructing a model structure corresponding to the preset model; The model structure includes: an attention layer, which determines the feature weight corresponding to each time in the historical indicator data; The data corresponding to each time period in the historical indicator data is respectively used as the first indicator data in chronological order; The feature weight is determined based on historical corrosion feature data corresponding to historical indicator data that is temporally adjacent to the first indicator data.

[0047] In addition, in addition to the attention layer, the preset model can also be provided with an input layer, an embedding layer, a feature extraction layer, and an output layer. In other embodiments, the preset model needs to be provided with the necessary attention layer, and there is no restriction on the structure of the model.

[0048] In this embodiment, the formula of the attention layer is:

[0049] Among them, T is the duration of the overall corrosion, t is the specific moment corresponding to the corrosion data, is the feature weight at time t, tanh is the activation function, is the corrosion characteristic data at time t, of course for The calculation formula is:

[0050] After the feature weights are calculated, they are passed to the feature extraction layer. This embodiment uses a deep learning model similar to the additive attention mechanism, which was first used in the neural machine translation (NMT) task. The original formula is:

[0051] in, is the context vector required to generate the i-th target word. is the annotation vector of the jth word in the source sentence, is the attention weight, which indicates the importance weight of the source word when generating the target word. is the length of the source sentence.

[0052]

[0053] in, Alignment score, which measures how well the source and target words match.

[0054] Subsequently, the additive attention mechanism is generally used in video data processing to assign corresponding weights to multiple image frames, allowing the model to focus on the video images of key frames, so that better time series information can be obtained. The core change to the original formula is to use corrosion feature data as the data for calculating weights. Referring to the second embodiment, since the corrosion feature data therein is normalized, it reflects the relationship with the corrosion jump value. Using corrosion feature data as data for calculating weights can effectively make the model focus on indicator data in a state of high corrosion rate.

[0055] After obtaining the feature weights, the input of the feature extraction layer here is:

[0056] in, is the sum of all input features under the overall corrosion time length, for the model training process is the historical indicator data at time t. In the step of determining the corrosion prediction result based on the corrosion characteristic data, indicator data and corrosion monitoring identification model, is the indicator data at time t.

[0057] The technical starting point of this embodiment is that, through the attention layer calculation method, the model's attention can be effectively distributed to indicators with high corrosion characteristics. Specifically, when all the input indicators are within the range that allows the material's oxide film to exist stably, the model's attention can be distributed more evenly to each indicator. In addition, the output of the feature extraction layer can be sent to the next processing layer of the model, which can be the output layer or the fully connected layer.

[0058] In this embodiment, the historical corrosion monitoring and recognition model calculates historical corrosion characteristic data using the historical indicator data and the material data, and generates corresponding training and validation sets based on the corrosion results, historical indicator data, and historical corrosion characteristic data. Furthermore, because the model structure includes an attention layer, the weights of each input data are determined in the attention layer and combined with the historical corrosion characteristic data. This allows the weights, i.e., attention, to be dynamically allocated to indicator data with high corrosion efficiency, thereby improving the accuracy of corrosion prediction results.

[0059] Furthermore, based on any of the above embodiments, a fifth embodiment of the multi-parameter water bath anti-corrosion monitoring method of the present invention is proposed, wherein the corrosion prediction result includes the corrosion thickness, and the step of determining the monitoring result based on the material thickness and the corrosion prediction result includes: The real-time material thickness of the material of the water bath is calculated according to the material thickness and the corrosion thickness.

[0060] In this embodiment, the real-time material thickness is obtained by subtracting the corrosion thickness from the material thickness, and the real-time material thickness is used as the monitoring result. The real-time material thickness is calculated by combining the predicted corrosion thickness with the material thickness, thereby realizing the monitoring of the corrosion of the water bath.

[0061] Furthermore, the step of obtaining the material data of the water bath and the index data of the target medium in the water bath includes: Determining the material data according to the model of the water bath; The sensor is controlled to detect the composition of the target medium in the water bath at a preset frequency to obtain the index data.

[0062] In this embodiment, the material data is determined by the model of the water bath, and the sensor is controlled to detect the composition of the target medium in the water bath at a preset frequency to obtain the index data, thereby improving the real-time performance and accuracy of the index data.

[0063] In addition, an embodiment of the present invention further provides a water bath anti-corrosion monitoring system, the water bath anti-corrosion monitoring system comprising: an acquisition module, configured to acquire material data of a water bath and index data of a target medium in the water bath, wherein the material data includes: material type, material thickness, and a corrosion rate jump value of the material type in contact with a corrosive medium; a calculation module, configured to determine corrosion characteristic data of the target medium based on the material data and the index data; an identification module, configured to determine a corrosion prediction result based on the corrosion characteristic data, the indicator data, and a corrosion monitoring identification model, wherein the corrosion monitoring identification model is provided with an attention layer, and a data weight corresponding to the indicator data is determined by the attention layer and the corrosion characteristic data; A monitoring module is used to determine a monitoring result based on the material thickness and the corrosion prediction result.

[0064] The water bath anti-corrosion monitoring system can implement the steps of the water bath anti-corrosion monitoring method of any of the above embodiments.

[0065] In addition, an embodiment of the present invention also proposes a water bath, which includes: a memory, a processor, and a water bath anti-corrosion monitoring program stored on the memory and runnable on the processor, wherein the water bath anti-corrosion monitoring program is configured to implement the steps of the water bath anti-corrosion monitoring method of any of the above embodiments.

[0066] In addition, an embodiment of the present invention further provides a storage medium storing a water bath anti-corrosion monitoring program. When the water bath anti-corrosion monitoring program is executed by a processor, the steps of the water bath anti-corrosion monitoring method of any of the above embodiments are implemented.

Claims

1. A multi-parameter water bath anti-corrosion monitoring method, characterized in that: The multi-parameter water bath anti-corrosion monitoring method comprises the following steps: Acquiring material data of a water bath and index data of a target medium in the water bath, wherein the material data includes: material type, material thickness, and a corrosion rate jump value of the material type when exposed to a corrosive medium; Determining corrosion characteristic data of the target medium according to the material data and the index data; Determining a corrosion prediction result based on the corrosion characteristic data, the indicator data, and a corrosion monitoring and identification model, wherein the corrosion monitoring and identification model is provided with an attention layer, and the data weight corresponding to the indicator data is determined by the attention layer and the corrosion characteristic data; A monitoring result is determined according to the material thickness and the corrosion prediction result.

2. The multi-parameter water bath anti-corrosion monitoring method according to claim 1, characterized in that: The corrosion rate jump value includes: a strong acid corrosion jump value, a strong alkali corrosion jump value, a chloride ion corrosion jump value and a corresponding critical temperature pitting corrosion value. The step of determining the corrosion characteristic data of the target medium according to the material data and the indicator data includes: Normalizing the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, the critical temperature pitting corrosion value, and the indicator data; Calculate the difference between the indicator data and the strong acid corrosion jump value, the strong alkali corrosion jump value, the chloride ion corrosion jump value, and the critical temperature pitting corrosion value to obtain a plurality of corrosion change differences; The corrosion characteristic data is determined according to a plurality of the corrosion change differences.

3. The multi-parameter water bath anti-corrosion monitoring method according to claim 2, characterized in that: The indicator data includes: pH value data, chloride ion concentration data and temperature data. The step of calculating the difference between the indicator data and the strong acid corrosion jump value, the strong base corrosion jump value, the chloride ion corrosion jump value and the critical temperature pitting corrosion value to obtain multiple corrosion change differences includes: Calculating a first corrosion change difference according to the pH value data and the strong acid corrosion jump value; Calculating a second corrosion change difference according to the pH value data and the strong alkali corrosion jump value; Calculating a third corrosion change difference according to the chloride ion concentration data and the chloride ion corrosion jump value; A fourth corrosion variation difference is calculated based on the temperature data and the critical temperature pitting value.

4. The multi-parameter water bath anti-corrosion monitoring method according to claim 1, characterized in that: Before the step of determining the corrosion prediction result based on the corrosion characteristic data, the index data and the corrosion monitoring identification model, the method further includes: Obtaining historical indicator data of the medium in the water bath before the current moment, and measuring corresponding corrosion results; Calculating historical corrosion characteristic data based on the historical indicator data and the material data; Determine and generate corresponding training sets and verification sets based on the corrosion results, historical indicator data, and historical corrosion characteristic data; Training is performed according to the training set and the preset model, and the trained model is verified according to the verification set to obtain the corrosion monitoring and recognition model.

5. The multi-parameter water bath anti-corrosion monitoring method according to claim 4, characterized in that: Before the step of training according to the training set and the preset model, the method further includes: Constructing a model structure corresponding to the preset model, wherein the model structure includes: an attention layer, wherein the attention layer determines the feature weight corresponding to each time in the historical indicator data; The data corresponding to each time period in the historical indicator data is respectively used as the first indicator data in chronological order; The feature weight is determined based on historical corrosion feature data corresponding to historical indicator data that is temporally adjacent to the first indicator data.

6. The multi-parameter water bath anti-corrosion monitoring method according to claim 1, characterized in that: The corrosion prediction result includes corrosion thickness, and the step of determining the monitoring result according to the material thickness and the corrosion prediction result includes: The real-time material thickness of the material of the water bath is calculated according to the material thickness and the corrosion thickness.

7. The multi-parameter water bath anti-corrosion monitoring method according to any one of claims 1 to 6, characterized in that: The step of obtaining material data of the water bath and index data of the target medium in the water bath comprises: Determining the material data according to the model of the water bath; The sensor is controlled to detect the composition of the target medium in the water bath at a preset frequency to obtain the index data.

8. A water bath anti-corrosion monitoring system, characterized in that: The water bath anti-corrosion monitoring system includes: an acquisition module, configured to acquire material data of a water bath and index data of a target medium in the water bath, wherein the material data includes: material type, material thickness, and a corrosion rate jump value of the material type in contact with a corrosive medium; a calculation module, configured to determine corrosion characteristic data of the target medium based on the material data and the index data; an identification module, configured to determine a corrosion prediction result based on the corrosion characteristic data, the indicator data, and a corrosion monitoring identification model, wherein the corrosion monitoring identification model is provided with an attention layer, and a data weight corresponding to the indicator data is determined by the attention layer and the corrosion characteristic data; A monitoring module is used to determine a monitoring result based on the material thickness and the corrosion prediction result.

9. A water bath, characterized in that: The water bath comprises: a memory, a processor, and a water bath anti-corrosion monitoring program stored in the memory and executable on the processor, wherein the water bath anti-corrosion monitoring program is configured to implement the steps of the water bath anti-corrosion monitoring method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a water bath anti-corrosion monitoring program, which, when executed by the processor, implements the steps of the water bath anti-corrosion monitoring method according to any one of claims 1 to 7.

Images