Descaling detection method and system for evaporator

Abstract

The invention relates to the technical field of dirt thickness detection, and discloses a descaling detection method and system for an evaporator, and the system comprises a dirt composition structural feature acquisition module, a dirt composition structural feature analysis module, and a descaling scheme formulation output module. Through a dirt composition structure feature acquisition module and a dirt composition structure feature analysis module, dirt composition structure detection image data of an axial cross section in an evaporator tube is collected, is subjected to data smoothing processing and noise reduction, and is matched with a dirt composition structure type image data algorithm to output dirt composition structure type feature data; therefore, accurate analysis of the dirt composition structure characteristics in the evaporator is realized, a specific descaling mode and a specific descaling process of each dirt layer are formed according to matching of the dirt composition structure type characteristic data, the thickness mean value and the descaling operation time of each dirt layer are measured, and the dirt composition structure type characteristic data are integrated; therefore, customization and scientization of descaling operation in the evaporator are realized.

Description

Technical Field

[0001] This invention relates to the technical field of dirt thickness detection, specifically to a descaling detection method and system for evaporators. Background Technology

[0002] An evaporator is an apparatus that transforms a liquid substance into a gaseous state. An evaporator mainly consists of two parts: a heating chamber and an evaporation chamber. The heating chamber provides the heat required for evaporation, causing the liquid to boil and vaporize; the evaporation chamber completely separates the gas and liquid phases. The vapor generated in the heating chamber carries a large amount of liquid droplets. Upon reaching the larger evaporation chamber, these liquid droplets are separated from the vapor through self-condensation or the action of a demister. Typically, the demister is located at the top of the evaporation chamber. The circulating cooling water inside the evaporator contains a large amount of salts, corrosion products, and various microorganisms. Due to the lack of water treatment, after the evaporator has been running for a period of time, a large amount of calcium and magnesium carbonate scale, algae, microbial sludge, and slime will form on the water side, firmly adhering to the inner surface of the copper tubes. Existing methods for cleaning the internal scale of evaporators simply involve using acidic descaling agents. However, due to changes in the quality of the circulating cooling water inside the evaporator, the scale composition becomes layered, resulting in poor cleaning effectiveness. Current evaporator internal scale cleaning processes lack pre-cleaning structural analysis and testing of the scale composition. They cannot develop descaling procedures and methods based on the structural characteristics of the internal scale, nor can they achieve online monitoring and feedback of the descaling process based on the thickness of the internal scale. This leads to problems such as over-cleaning, damaging the evaporator piping structure, or incomplete cleaning.

[0003] Chinese Patent Publication No. CN111140837A discloses a descaling control method and system for a steam generator. This method involves supplying a descaling agent solution to the steam generator at a low flow rate, using chemical soaking to break the adhesion between the scale and the steam generator. Then, a high flow rate of the descaling agent solution is supplied to the steam generator, using high-speed impact to knock off the scale and chemically dissolve it, reducing descaling time. The descaling system of this invention uses a controller to control the working power of the water pump and the opening degree of the solenoid valve to easily control the liquid supply flow rate and achieve variable-speed flushing. This descaling method, combining low-flow-rate descaling agent soaking and high-flow-rate flushing, is effective for single-component scale. However, for scale with layered structures, existing evaporator internal scale cleaning processes lack pre-cleaning structural analysis and detection. They cannot formulate descaling processes and methods based on the structural characteristics of the scale inside the evaporator, nor can they achieve online monitoring and feedback of the descaling process based on the scale thickness. This can lead to over-cleaning that damages the evaporator piping structure or incomplete cleaning. Summary of the Invention

[0004] (a) Technical problems to be solved To address the shortcomings of existing methods that simply use acidic descaling agents to clean internal evaporator scale, which suffer from poor cleaning results due to variations in the circulating cooling water quality and the resulting stratified scale composition, current evaporator cleaning processes lack pre-cleaning structural analysis of the scale. This prevents the development of descaling procedures and methods based on the structural characteristics of the scale, and also hinders online monitoring and feedback of the descaling process based on scale thickness. Consequently, over-cleaning can damage evaporator piping or result in incomplete cleaning. The goal is to achieve precise analysis of scale composition, customized descaling procedures and methods, online monitoring of the descaling process, and accurate and efficient descaling.

[0005] (II) Technical Solution This invention is achieved through the following technical solution: a method for detecting descaling in an evaporator, comprising the following steps: S1. Detect and acquire image data of the fouling composition structure of the axial cross section inside the evaporator tube; S2. Data smoothing and noise reduction are applied to the image data of the dirt composition structure detection to generate image data of the dirt cross-sectional composition structure features; S3. Obtain image data of the structural features of the cross-section of the dirt and image data of the types of dirt composition structure. Use an algorithm model to identify and output the structural features of the dirt composition structure from the axis of the evaporator tube to the inner diameter circumference. S4. Traverse the data on the composition and structural characteristics of dirt to match and form the descaling method and process; measure the average thickness of each dirt layer in the image data of the composition and structural characteristics of dirt cross-section, and measure the descaling operation time based on the average thickness; integrate the descaling method, process and descaling time to form the output descaling operation plan data. S5. After the descaling operation, the average thickness of the dirt layer in the real-time measured dirt cross-section structural feature image data is used to make a judgment. If the average thickness of the dirt layer reaches the set dirt layer thickness threshold, the descaling operation will continue until the descaling operation is completed. Otherwise, the descaling operation will be repeated. S6. Based on the descaling operation results in S5, provide online feedback output.

[0006] Preferably, the operation steps for detecting and acquiring the fouling composition structure detection image data of the axial cross-section inside the evaporator tube are as follows: S11. Use an X-ray detector to capture image data of the fouling structure in the axial section of the evaporator. S12. Establish a detection image data matrix for the composition and structure of dirt; , ,in Indicates the first Image data of the structure of dirt.

[0007] Preferably, the steps for generating cross-sectional structural feature image data of dirt by performing data smoothing and noise reduction on the detected dirt composition structure image data are as follows: S21, Call the image data matrix of dirt composition structure detection Image data of dirt composition and structure detection ; S22. The exponential moving average method is used to detect the image data matrix of dirt composition structure. Smoothing and noise reduction processing is performed; the exponential moving average method uses the following formula; ,in, Indicates the predicted value. Indicates the decay weight. Represents the observed value; S23. Generating a matrix of image data featuring the cross-sectional structure of dirt using smoothing and noise reduction processing. , ,in Indicates the first Image data of structural features composed of cross-sections of dirt. Preferably, the steps for obtaining the image data of the structural features of the dirt cross-section and the image data of the types of dirt composition and structure using an algorithm model to identify and output the characteristic data of the types of dirt composition and structure in the direction from the axis of the evaporator tube to the inner diameter circumference are as follows: S31. Establish a set of image data on the composition and structure of dirt. , ,in Indicates the first Image data of the composition and structure of each type of dirt; S32, Call the structural feature image data matrix of the dirt cross-section. Using water wave algorithms and image datasets of dirt composition and structure types The steps for identification and matching are as follows: Propagation: Image data of the structural features of dirt cross-section If we define it as water ripples, with each ripple being an independent entity, then each ripple will have three attributes: position. ,wavelength and wave height ; In each iteration, each water wave propagates through a collection of image data on the composition and structure of the dirt. Perform search matching while water wave height It will decrease by 1, and its position update formula is as follows: ,in , These are the upper and lower bounds of the current search space, respectively. The value will change as the iteration proceeds: , of which Attenuation coefficient of wavelength, Use a smaller number to ensure the denominator is not zero; Refraction: in a water wave After propagation, the water wave may refract and... To identify the height of the water wave during each propagation. It will decrease by 1 when When the water wave decreases to 0, it will refract, and its height and wavelength will also change. The formulas for refraction and changes in height and wavelength are as follows: The refracted positions are normally distributed at locations with the midpoint between the current wave and the optimal wave as the mean, and the distance between the current wave and the optimal wave as the variance. The height of the refracted wave will be reinitialized to its maximum height. After refraction, The wavelength of the water wave will be recalculated. ; Breaking Waves: In the ripples of water Image data set of dirt composition and structure types After propagating internally, it reaches a position superior to the current optimal water wave, meaning it has searched and matched the optimal set of image data on the composition and structure of the dirt. The water wave will then break up and propagate the current optimal water wave to the location where the breakup occurs, thus outputting the structural feature image data of the dirt cross-section. Image data of optimal dirt composition and structure types ; The formula for determining the location of the wave breakers is as follows: , It is a random number, and each time the waves break, it will be randomly selected. Changes can be made in several dimensions. It is a constant; S33 and S32 use a water wave algorithm to identify and match the output image data matrix of the structural features of the dirt cross-section. Image data of corresponding dirt composition and structure types .

[0008] Preferably, the operation steps for matching the characteristic data of the composition and structure of dirt to form a descaling method and process are as follows: S41, Call the image data matrix of the structural features of the dirt cross-section in S33. Image data of corresponding dirt composition and structure types ; S42. Establish a set of image data on the composition and structure of dirt. Data set of corresponding dirt composition, structural types, characteristics, descaling methods, and processes. , ,in Indicates the first Data on the composition, structure, types, characteristics, descaling methods, and processes of individual dirt deposits; S43, Based on the image data matrix of the structural characteristics of the dirt cross-section in S33 Image data of corresponding dirt composition and structure types Data set on the composition, structure, types, characteristics, descaling methods, processes, and techniques of dirt. Matching output of the structural feature image data matrix of dirt cross-section Corresponding dirt composition, structural types, characteristics, descaling methods, and process data. .

[0009] Preferably, the average thickness of each dirt layer in the image data of the structural features of the dirt cross-section is measured, and the descaling operation time is measured based on the average thickness; the operation steps for integrating the descaling method, process, and descaling time to form the output descaling operation plan data are as follows: S51, Image data matrix based on the structural characteristics of dirt cross-section Composition, structure, type, characteristics, descaling methods, process data Solve the image data matrix of the structural features of the fouling cross-section along the circumference of the evaporator tube axis from the tube axis to the inner diameter. The mean thickness of each dirt layer is calculated and a matrix of mean dirt layer thickness is constructed. , ,in Indicates the first circumference along the axis of the evaporator tube from the inner diameter circumference. Average thickness of various types of fouling layers; common evaporator fouling layers include one or all of the following: calcium magnesium carbonate scale, algae scale, microbial sludge scale, and slime scale. , Indicates the first The structural feature image data of the dirt cross-section, along the direction from the evaporator tube axis to the inner diameter circumference, is the first... The thickness of the dirt layer, ; S52. Solve for the descaling operation time of each fouling layer and establish a set. , ; Indicates the first circumference along the axis of the evaporator tube from the inner diameter circumference. The descaling operation time for this type of dirt layer; among which The unit is s. Indicates the first The descaling rate of each type of fouling layer using a corresponding descaling agent under certain temperature conditions is expressed in mm / s; among them, calcium magnesium carbonate scale is treated with acidic descaling agent, algae scale with protease descaling agent, microbial sludge scale with alkaline descaling agent, and slime scale with neutral descaling agent. S53, Image data matrix based on the structural characteristics of dirt cross-section Image data of the composition, structure, and types of dirt. Fouling composition, structure, types, characteristics, descaling methods, process data Collection of descaling operation times for each layer of dirt The system identifies and measures a descaling operation plan that includes image data of the types of dirt composition and structure, data on the descaling methods and processes based on the characteristics of the dirt composition and structure, and the descaling operation time for each dirt layer.

[0010] Preferably, after the descaling operation, the average thickness of the dirt layer in the real-time measured cross-sectional structural feature image data is used to determine whether the average dirt layer thickness reaches a set dirt layer thickness threshold. If so, the descaling operation continues until it is completed; otherwise, the operation is repeated as follows: S61. Implement the descaling operation plan in S53, and perform descaling operations according to the time required for each scale layer. Perform descaling operations on the evaporator tube walls in an orderly manner; S62. Determine the time limit for descaling operations on each layer of fouling. Complete the first step in an orderly manner After the descaling operation of the first type of dirt layer, obtain and determine the first... Average thickness of the dirt layer after descaling operation With the Average threshold for descaling thickness of various types of dirt layers relation; when ≤ , indicating the first The descaling operation for this type of fouling layer met the descaling requirements; therefore, the descaling operation plan should continue to be implemented. The descaling operation is carried out on various types of dirt layers until the descaling operation plan is completed. when > , indicating the first The descaling operation for this type of fouling layer does not meet the descaling requirements, and the descaling operation plan needs to be repeated. Descaling operations for various types of dirt layers.

[0011] Preferably, the operation steps for online feedback output based on the descaling operation results in S5 are as follows: S71, After descaling the dirt layer according to S62 and The system provides real-time feedback on the descaling results for each layer of dirt. In each stage of the descaling process, when ≤ Output the first Results of descaling operation for various types of dirt layers; when > Output the first The descaling operation for this type of dirt layer was not completed. When the descaling operation plan is completed, output the descaling operation plan completion result; A system for implementing the aforementioned descaling detection method for evaporators, the system comprising a fouling composition and structure feature acquisition module, a fouling composition and structure feature analysis module, a descaling scheme formulation and output module, and an online control feedback module for the descaling process; The fouling composition structure feature acquisition module includes a fouling cross-sectional composition structure detection unit and a fouling cross-sectional composition structure feature image generation unit; the fouling cross-sectional composition structure detection unit uses an X-ray detector to detect and acquire fouling composition structure detection image data of the axial cross-section inside the evaporator tube; the fouling cross-sectional composition structure feature image generation unit uses data smoothing processing to reduce noise from the fouling composition structure detection image data to generate fouling cross-sectional composition structure feature image data. The fouling composition and structure feature analysis module includes a fouling composition and structure type feature image storage unit and a fouling composition and structure type feature recognition output unit. The fouling composition and structure type feature image storage unit is used to store fouling composition and structure type image data. The fouling composition and structure type feature recognition output unit acquires fouling cross-sectional composition and structure feature image data and fouling composition and structure type image data, and uses an algorithm model to identify and output fouling composition and structure type feature data from the evaporator tube axis to the inner diameter circumference. The descaling solution output module includes a descaling process and method matching unit, a descaling operation time measurement unit, and a descaling operation overall solution output unit. The descaling process and method matching unit matches the characteristics of the fouling composition structure to form the descaling method and process for each fouling layer. The descaling operation time measurement unit measures the average thickness of each fouling layer in the fouling cross-sectional composition structure characteristic image data and measures the descaling operation time based on the average thickness. The descaling operation overall solution output unit integrates the characteristics of the fouling composition structure, the descaling method and process, and the descaling time to form an output descaling operation solution. The online control feedback module for the descaling process includes a unit for identifying the presence of structural features of the descaling composition and a unit for detecting and judging the descaling result. The unit for identifying the presence of structural features of the descaling composition is used to judge and identify the descaling result based on the average thickness of the descaling layer measured in the real-time acquired image data of the structural features of the descaling cross-section. The unit for detecting and judging the descaling result provides online feedback output on the descaling result.

[0012] (III) Beneficial Effects This invention provides a method and system for detecting descaling in evaporators. It offers the following advantages: The system employs a fouling composition and structure feature acquisition module and a fouling composition and structure feature analysis module to collect fouling composition and structure detection image data of the axial cross-section inside the evaporator tubes. This data is then smoothed and noise-reduced, and matched with fouling composition and structure type image data using an algorithm to output fouling composition and structure type feature data, thereby achieving accurate analysis of the fouling composition and structure features inside the evaporator. The descaling scheme formulation and output module matches the fouling composition and structure type feature data to form specific descaling methods and processes for each fouling layer, measures the average thickness of each fouling layer and the descaling operation time, and integrates the fouling composition and structure type feature data, descaling methods and processes, and descaling time to form an output descaling operation plan. This achieves customized and scientific descaling operations inside the evaporator. The online control and feedback module for the descaling process is used to judge and identify the fouling layer thickness based on the measured average thickness from the real-time acquired fouling cross-sectional composition and structure feature image data after the descaling operation, and provides online feedback output of the descaling operation results. This achieves online result feedback monitoring of the descaling process inside the evaporator, improving the efficiency and quality of descaling inside the evaporator.

[0013] Second, the descaling process and method matching unit uses the characteristics of the fouling composition structure to match the descaling method process data and outputs the fouling cross-sectional composition structure characteristic image data corresponding to the fouling composition structure type and descaling method process data. This makes the descaling process inside the evaporator more accurate and scientific. At the same time, the descaling operation time measurement unit establishes a matrix of the average thickness of each fouling layer. The ratio of the average thickness of each type of fouling layer to the descaling rate of the corresponding descaling agent is used to calculate the descaling operation time of each fouling layer. This enables quantitative control of the descaling operation time inside the evaporator and increases the quality of descaling.

[0014] Third, by identifying the structural characteristics of the fouling after descaling, the system determines the fouling layer and limits the descaling operation time for each fouling layer. After the descaling operation is completed, the average thickness of the fouling layer after the operation is completed is obtained online and compared with the corresponding average thickness threshold of the fouling layer. The system monitors the descaling results of each fouling layer, forming a closed-loop online monitoring of the descaling process inside the evaporator, ensuring the effectiveness of descaling inside the evaporator. The descaling result detection and judgment unit provides online feedback output of the descaling operation results, improving the understanding and collection of information on the descaling process results inside the evaporator, and achieving accurate monitoring of descaling inside the evaporator. Attached Figure Description

[0015] Figure 1 This invention provides a diagram showing the composition of various modules of a descaling and detection system for an evaporator. Figure 2 for Figure 1 The diagram shows the operational structure of a descaling detection method for evaporators. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] An embodiment of the descaling detection method and system for evaporators is as follows: Please see Figures 1-2 A method for detecting descaling in evaporators, comprising the following steps: S1. Detect and acquire image data of the fouling composition structure of the axial cross section inside the evaporator tube; S2. Data smoothing and noise reduction are applied to the image data of the dirt composition structure detection to generate image data of the dirt cross-sectional composition structure features; S3. Obtain image data of the structural features of the cross-section of the dirt and image data of the types of dirt composition structure. Use an algorithm model to identify and output the structural features of the dirt composition structure from the axis of the evaporator tube to the inner diameter circumference. S4. Traverse the data on the composition and structural characteristics of dirt to match and form the descaling method and process; measure the average thickness of each dirt layer in the image data of the composition and structural characteristics of dirt cross-section, and measure the descaling operation time based on the average thickness; integrate the descaling method, process and descaling time to form the output descaling operation plan data. S5. After the descaling operation, the average thickness of the dirt layer in the real-time measured dirt cross-section structural feature image data is used to make a judgment. If the average thickness of the dirt layer reaches the set dirt layer thickness threshold, the descaling operation will continue until the descaling operation is completed. Otherwise, the descaling operation will be repeated. S6. Based on the descaling operation results in S5, provide online feedback output.

[0018] For further details, please refer to Figures 1-2 The operation steps for detecting and acquiring image data of the fouling composition structure of the axial cross-section inside the evaporator tube are as follows: S11. Use an X-ray detector to capture image data of the fouling structure in the axial section of the evaporator. S12. Establish a detection image data matrix for the composition and structure of dirt; , ,in Indicates the first Image data of the structure of dirt.

[0019] The steps for generating cross-sectional structural feature image data of dirt by performing data smoothing and noise reduction on the image data of dirt composition and structure detection are as follows: S21, Call the image data matrix of dirt composition structure detection Image data of dirt composition and structure detection ; S22. The exponential moving average method is used to detect the image data matrix of dirt composition structure. Smoothing and noise reduction processing is performed; the exponential moving average method uses the following formula; ,in, Indicates the predicted value. Indicates the decay weight. Represents the observed value; S23. Generating a matrix of image data featuring the cross-sectional structure of dirt using smoothing and noise reduction processing. , ,in Indicates the first Image data of structural features composed of cross-sections of dirt.

[0020] The steps for obtaining image data of the structural features of the cross-section of the fouling and image data of the types of fouling structural components, and using an algorithm model to identify and output the structural features of the fouling in the direction from the axis of the evaporator tube to the inner diameter circumference, are as follows: S31. Establish a set of image data on the composition and structure of dirt. , ,in Indicates the first Image data of the composition and structure of each type of dirt; S32, Call the structural feature image data matrix of the dirt cross-section. Using water wave algorithms and image datasets of dirt composition and structure types The steps for identification and matching are as follows: Propagation: Image data of the structural features of dirt cross-section If we define it as water ripples, with each ripple being an independent entity, then each ripple will have three attributes: position. ,wavelength and wave height ; In each iteration, each water wave propagates through a collection of image data on the composition and structure of the dirt. Perform search matching while water wave height It will decrease by 1, and its position update formula is as follows: ,in , These are the upper and lower bounds of the current search space, respectively. The value will change as the iteration proceeds: , of which Attenuation coefficient of wavelength, Use a smaller number to ensure the denominator is not zero; Refraction: in a water wave After propagation, the water wave may refract and... To identify the height of the water wave during each propagation. It will decrease by 1 when When the water wave decreases to 0, it will refract, and its height and wavelength will also change. The formulas for refraction and changes in height and wavelength are as follows: The refracted positions are normally distributed at locations with the midpoint between the current wave and the optimal wave as the mean, and the distance between the current wave and the optimal wave as the variance. The height of the refracted wave will be reinitialized to its maximum height. After refraction, The wavelength of the water wave will be recalculated. ; Breaking Waves: In the ripples of water Image data set of dirt composition and structure types After propagating internally, it reaches a position superior to the current optimal water wave, meaning it has searched and matched the optimal set of image data on the composition and structure of the dirt. The water wave will then break up and propagate the current optimal water wave to the location where the breakup occurs, thus outputting the structural feature image data of the dirt cross-section. Image data of optimal dirt composition and structure types ; The formula for determining the location of the wave breakers is as follows: , It is a random number, and each time the waves break, it will be randomly selected. Changes can be made in several dimensions. It is a constant; S33 and S32 use a water wave algorithm to identify and match the output image data matrix of the structural features of the dirt cross-section. Image data of corresponding dirt composition and structure types .

[0021] For further details, please refer to Figures 1-2 The operational steps for matching and determining the descaling method and process by traversing the data on the composition, structure, type, and characteristics of dirt are as follows: S41, Call the image data matrix of the structural features of the dirt cross-section in S33. Image data of corresponding dirt composition and structure types ; S42. Establish a set of image data on the composition and structure of dirt. Data set of corresponding dirt composition, structural types, characteristics, descaling methods, and processes. , ,in Indicates the first Data on the composition, structure, types, characteristics, descaling methods, and processes of individual dirt deposits; S43, Based on the image data matrix of the structural characteristics of the dirt cross-section in S33 Image data of corresponding dirt composition and structure types Data set on the composition, structure, types, characteristics, descaling methods, processes, and techniques of dirt. Matching output of the structural feature image data matrix of dirt cross-section Corresponding dirt composition, structural types, characteristics, descaling methods, and process data. .

[0022] The steps for measuring the average thickness of each dirt layer in the cross-sectional structural feature image data, and then using this average thickness to measure the descaling operation time, are as follows: Integrating the descaling method, process, and time to generate the output descaling operation plan data. S51, Image data matrix based on the structural characteristics of dirt cross-section Composition, structure, type, characteristics, descaling methods, process data Solve the image data matrix of the structural features of the fouling cross-section along the circumference of the evaporator tube axis from the tube axis to the inner diameter. The mean thickness of each dirt layer is calculated and a matrix of mean dirt layer thickness is constructed. , ,in Indicates the first circumference along the axis of the evaporator tube from the inner diameter circumference. Average thickness of various types of fouling layers; common evaporator fouling layers include one or all of the following: calcium magnesium carbonate scale, algae scale, microbial sludge scale, and slime scale. , Indicates the first The structural feature image data of the dirt cross-section, along the direction from the evaporator tube axis to the inner diameter circumference, is the first... The thickness of the dirt layer, ; S52. Solve for the descaling operation time of each fouling layer and establish a set. , ; Indicates the first circumference along the axis of the evaporator tube from the inner diameter circumference. The descaling operation time for this type of dirt layer; among which The unit is s. Indicates the first The descaling rate of each type of fouling layer using a corresponding descaling agent under certain temperature conditions is expressed in mm / s; among them, calcium magnesium carbonate scale is treated with acidic descaling agent, algae scale with protease descaling agent, microbial sludge scale with alkaline descaling agent, and slime scale with neutral descaling agent. S53, Image data matrix based on the structural characteristics of dirt cross-section Image data of the composition, structure, and types of dirt. Fouling composition, structure, types, characteristics, descaling methods, process data Collection of descaling operation times for each layer of dirt The system identifies and measures a descaling operation plan that includes image data of the types of dirt composition and structure, data on the descaling methods and processes based on the characteristics of the dirt composition and structure, and the descaling operation time for each dirt layer.

[0023] The descaling process and method matching unit uses the characteristics of the fouling composition and structure to match the descaling method process data and outputs the corresponding fouling cross-sectional composition and structure image data. This makes the descaling process inside the evaporator more accurate and scientific. At the same time, the descaling operation time measurement unit establishes a matrix of the average thickness of each fouling layer. The ratio of the average thickness of each type of fouling layer to the descaling rate of the corresponding descaling agent is used to calculate the descaling operation time of each fouling layer. This enables quantitative control of the descaling operation time inside the evaporator and improves the quality of descaling.

[0024] For further details, please refer to Figures 1-2After the descaling operation, the average thickness of the dirt layer is determined based on the real-time measured cross-sectional structural feature image data. If the average dirt layer thickness reaches the set dirt layer thickness threshold, the descaling operation continues until it is completed; otherwise, the descaling operation is repeated. The operation steps are as follows: S61. Implement the descaling operation plan in S53, and perform descaling operations according to the time required for each scale layer. Perform descaling operations on the evaporator tube walls in an orderly manner; S62. Determine the time limit for descaling operations on each layer of fouling. Complete the first step in an orderly manner After the descaling operation of the first type of dirt layer, obtain and determine the first... Average thickness of the dirt layer after descaling operation With the Average threshold for descaling thickness of various types of dirt layers relation; when ≤ , indicating the first The descaling operation for this type of fouling layer met the descaling requirements; therefore, the descaling operation plan should continue to be implemented. The descaling operation is carried out on various types of dirt layers until the descaling operation plan is completed. when > , indicating the first The descaling operation for this type of fouling layer does not meet the descaling requirements, and the descaling operation plan needs to be repeated. Descaling operations for various types of dirt layers.

[0025] The steps for online feedback output based on the descaling operation results in S5 are as follows: S71, After descaling the dirt layer according to S62 and The system provides real-time feedback on the descaling results for each layer of dirt. In each stage of the descaling process, when ≤ Output the first Results of descaling operation for various types of dirt layers; when > Output the first The descaling operation for this type of dirt layer was not completed. When the descaling operation plan is completed, output the descaling operation plan completion result; The system identifies the structural characteristics of the fouling layer after descaling. After descaling is completed within a specified timeframe, the average thickness of the fouling layer after descaling is compared with a corresponding threshold value. This monitoring of each fouling layer's descaling results creates a closed-loop online monitoring system, ensuring effective descaling within the evaporator. The descaling result detection and judgment unit provides online feedback on the descaling results, improving the understanding and collection of information regarding the descaling process within the evaporator and enabling precise monitoring of descaling within the evaporator.

[0026] A system for implementing a descaling detection method for evaporators, the system comprising a fouling composition and structure feature acquisition module, a fouling composition and structure feature analysis module, a descaling scheme formulation and output module, and an online control feedback module for the descaling process; The fouling composition structure feature acquisition module includes a fouling cross-sectional composition structure detection unit and a fouling cross-sectional composition structure feature image generation unit. The fouling cross-sectional composition structure detection unit uses an X-ray detector to detect and acquire fouling composition structure detection image data of the axial cross-section inside the evaporator tube. The fouling cross-sectional composition structure feature image generation unit uses data smoothing processing to reduce noise from the fouling composition structure detection image data to generate fouling cross-sectional composition structure feature image data. The fouling composition and structure feature analysis module includes a fouling composition and structure type feature image storage unit and a fouling composition and structure type feature recognition output unit. The fouling composition and structure type feature image storage unit is used to store fouling composition and structure type image data. The fouling composition and structure type feature recognition output unit acquires fouling cross-sectional composition and structure feature image data and fouling composition and structure type image data, and uses an algorithm model to identify and output fouling composition and structure type feature data from the evaporator tube axis to the inner diameter circumference. The descaling solution output module includes a descaling process and method matching unit, a descaling operation time measurement unit, and a descaling operation overall solution output unit. The descaling process and method matching unit matches the descaling method and process for each fouling layer with the fouling composition and structural characteristics data. The descaling operation time measurement unit measures the average thickness of each fouling layer in the fouling cross-sectional composition and structural characteristics image data, and measures the descaling operation time based on the average thickness. The descaling operation overall solution output unit integrates the fouling composition and structural characteristics data, descaling methods and processes, and descaling time to form an output descaling operation solution. The online control feedback module for the descaling process includes a unit for identifying the presence of the structural features of the descaling composition and a unit for detecting and judging the descaling results. The unit for identifying the presence of the structural features of the descaling composition is used to judge and identify the descaling results based on the average thickness of the descaling layer measured in the real-time acquired image data of the structural features of the descaling cross-section. The unit for detecting and judging the descaling results provides online feedback output on the descaling results.

[0027] The system employs two modules: a fouling composition and structure feature acquisition module and a fouling composition and structure feature analysis module. These modules collect fouling composition and structure detection images of the axial cross-section inside the evaporator tubes. After data smoothing and noise reduction, the images are matched with fouling composition and structure type image data using an algorithm to output fouling composition and structure type feature data. This enables precise analysis of the fouling composition and structure characteristics inside the evaporator. A descaling scheme formulation and output module matches the fouling composition and structure type feature data to formulate specific descaling methods and processes for each fouling layer. It measures the average thickness of each fouling layer and the descaling operation time, integrating the fouling composition and structure type feature data, descaling methods and processes, and descaling time to generate an output descaling operation plan. This achieves customized and scientific descaling operations inside the evaporator. An online control and feedback module for the descaling process is used to identify and judge the fouling layer thickness based on the measured average thickness from the real-time acquired fouling cross-sectional composition and structure feature image data after the descaling operation. It also provides online feedback output of the descaling operation results, enabling online monitoring of the descaling process inside the evaporator and improving the efficiency and quality of descaling inside the evaporator.

Claims

1. A method for scale detection for an evaporator, characterized in that, It comprises the following steps: S1, detecting and acquiring fouling composition structure detection image data of the axial section inside the evaporator tube; S2, generating fouling section composition structure feature image data by data smoothing processing and noise reduction on the fouling composition structure detection image data; S3, acquiring fouling section composition structure feature image data and fouling composition structure type image data, and identifying and outputting fouling composition structure type feature data of the evaporator tube axis to the inner diameter circumferential direction by using an algorithm model; S4, traversing the fouling composition structure type feature data to match and form a descaling mode and process; measuring the thickness average of each fouling layer in the fouling section composition structure feature image data, and performing descaling operation time measurement according to the thickness average; integrating the descaling mode and process and the descaling time to form output descaling operation scheme data; S5, after the descaling operation, judging according to the fouling layer thickness average in the real-time measured fouling section composition structure feature image data; if the fouling layer thickness average reaches the set fouling layer thickness threshold, the descaling operation is continued until the descaling operation is completed, otherwise the descaling operation is restarted; S6, online feedback output according to the descaling operation result in S5.

2. A method for scale detection in an evaporator according to claim 1, characterized in that: The operation steps of detecting and acquiring fouling composition structure detection image data of the axial section inside the evaporator tube are as follows: S11, using an X-ray detector to take pictures to generate fouling composition structure detection image data of the axial section of the evaporator; S12, a fouling composition structure detection image data matrix is established; , wherein represents the th fouling composition structure detection image data.

3. A method of fouling detection for an evaporator as claimed in claim 2, wherein: The operation steps of generating fouling section composition structure feature image data by data smoothing processing and noise reduction on the fouling composition structure detection image data are as follows: S21, calling the fouling composition structure detection image data matrix middle fouling composition structure detection image data ; S22, using an exponential moving average method to detect the image data matrix of the fouling composition structure Smooth noise reduction processing is performed; where the exponential smoothing method employs the formula where, denotes the predicted value, denotes the decay weight, denotes the observed value; S23, generating the fouling cross-section composition structure feature image data matrix by using the smooth noise reduction processing , wherein represents the i th fouling cross-section composition structure feature image data.

4. A method of fouling detection for an evaporator as claimed in claim 3, wherein: The operation steps of acquiring fouling section composition structure feature image data and fouling composition structure type image data, and identifying and outputting fouling composition structure type feature data of the evaporator tube axis to the inner diameter circumferential direction by using an algorithm model are as follows: S31, establish a set of dirt composition structure type image data , wherein represents the th dirt composition structure type image data; S32, calling fouling cross-section composition structure feature image data matrix using a water wave algorithm and a fouling composition structure type image data set The operation steps of the identification matching are as follows: Propagation: dirt cross-sectional composition structure feature image data Set to water waves, each water wave is an independent individual, then each water wave will have three attributes: position , wavelength and wave height ; In each iteration, each water wave propagates through the set of images of the structure of the dirt composition The search matching while the height of the water wave Will reduce 1, its position update formula is as follows: where , are the upper and lower bounds of the current search space, respectively, the values of will change as the iterations proceed: wherein a is the attenuation coefficient of the wavelength, is a small number to ensure that the denominator is not zero; Refraction: at a water wave After propagation, the water wave has the possibility to refract and Identify, each time the water wave propagates, the height of the water wave will decrease by 1, when The water wave will refract when the height is reduced to 0, and the height and wavelength will also change, the refractive and height wavelength change formula is as follows: ； The position after refraction is normally distributed in the position with the current water wave and the midpoint of the optimal water wave as the mean value, and the distance between the current water wave and the optimal water wave as the variance. The height of the water wave after refraction will be reinitialized as the maximum height. ; after refraction, The wavelength of the water wave will be recalculated: ; breaker: in water wave in the dirt composition structure kind image data set After the propagation in the dirt composition structure kind image data set, a position superior to the current optimal water wave is reached, i.e. the search matches to the optimal dirt composition structure kind image data set , the water wave will break and the current optimal water wave will be propagated to the position where the breakers are generated, i.e. the output is the dirt cross-section composition structure feature image data optimal dirt composition structure kind image data ; The formula for generating broken wave position is as follows: , is a random number, each time the surf will randomly select dimensions to change, is a constant; S33, output S32 in the water wave algorithm recognition matching output dirt cross-section composition structure feature image data matrix corresponding dirt composition structure category image data .

5. A method of scale detection for an evaporator as claimed in claim 4, wherein: The operation steps of traversing the fouling composition structure type feature data to match and form a descaling mode and process are as follows: S41, calling the dirt cross-section composition structure feature image data matrix in S33 corresponding dirt composition structure category image data ; S42, establish a set of dirt composition structure type image data corresponding dirt composition structure type feature cleaning method flow data set , wherein represents the th dirt composition structure type feature cleaning method flow data; S43, the fouling cross-section composition structure feature image data matrix according to S33 corresponding fouling composition structure type image data fouling composition structure type feature cleaning mode process data set matching output fouling cross-section composition structure feature image data matrix corresponding fouling composition structure type feature cleaning mode process data .

6. A method of scale detection for an evaporator as claimed in claim 5, wherein: The operation steps of measuring the thickness average of each fouling layer in the fouling section composition structure feature image data, and performing descaling operation time measurement according to the thickness average; integrating the descaling mode and process and the descaling time to form output descaling operation scheme data are as follows: S51, the fouling cross-section composition structure characteristic image data matrix according to the fouling cross-section composition structure characteristic image data matrix and the fouling composition structure kind characteristic cleaning mode flow data , solving the fouling cross-section composition structure characteristic image data matrix along the evaporator tube axis to the inner diameter circumferential direction The average thickness of each fouling layer in the matrix is calculated and the matrix of the average thickness of each fouling layer is established , , wherein represents the average thickness of the first fouling layer along the evaporator tube axis to the inner diameter circumferential direction; common evaporator fouling layers include one or all of calcium-magnesium carbonate scale, algal scale, microbial sludge scale, and viscous sludge scale; , represents the first fouling layer thickness in the first fouling cross-section composition structure characteristic image data along the evaporator tube axis to the inner diameter circumferential direction ; S52. Solve for the descaling operation time of each fouling layer and establish a set. , ; Indicates the first circumference along the axis of the evaporator tube from the inner diameter circumference. The descaling operation time for this type of dirt layer; among which The unit is s. Indicates the first The descaling rate of each type of fouling layer using a corresponding descaling agent under certain temperature conditions is expressed in mm / s; among them, calcium magnesium carbonate scale is treated with acidic descaling agent, algae scale with protease descaling agent, microbial sludge scale with alkaline descaling agent, and slime scale with neutral descaling agent. S53, a matrix of image data of structural features of the cross-section of the dirt , a matrix of image data of structural features of the cross-section of the dirt , a matrix of image data of structural features of the cross-section of the dirt , a matrix of image data of structural features of the cross-section of the dirt The scale of the dirt removal operation is identified by the matrix of image data of structural features of the cross-section of the dirt, the matrix of image data of structural features of the cross-section of the dirt, and the matrix of image data of structural features of the cross-section of the dirt.

7. A method of scale detection for an evaporator as claimed in claim 6, wherein: The operation steps of judging according to the fouling layer thickness average in the real-time measured fouling section composition structure feature image data after the descaling operation; if the fouling layer thickness average reaches the set fouling layer thickness threshold, the descaling operation is continued until the descaling operation is completed, otherwise the descaling operation is restarted are as follows: S61, performing the descaling operation scheme in S53 according to the descaling operation time of each scale layer The descaling operation is orderly performed on the evaporator tube wall; S62, judging the time of each scale layer cleaning operation after the orderly completion of the first scale layer cleaning operation, obtaining and judging the average thickness of the scale layer after the completion of the first scale layer cleaning operation and the average scale layer cleaning thickness threshold of the first scale layer relationship; When ≤ , it indicates that the first scale removal operation of the scale layer meets the scale removal requirement, and the second scale removal operation of the scale layer in the scale removal operation scheme is continued to be executed until the scale removal operation scheme is completed. ​ when > , indicating the first The descaling operation for this type of fouling layer does not meet the descaling requirements, and the descaling operation plan needs to be repeated. Descaling operations for various types of dirt layers.

8. A method of scale detection for an evaporator as claimed in claim 7, wherein: The operation steps of online feedback output according to the descaling operation result in S5 are as follows: S71、Deposits are removed according to S62 With relationship, real-time feedback of each layer of dirt removal work results; In each dirt layer descaling operation link, when ≤ , output the first dirt layer descaling operation completion result; when > , output the second dirt layer descaling operation incomplete result; When the descaling operation scheme is completed, the descaling operation scheme completion result is outputted; A system implementing a method for detecting fouling of an evaporator according to any one of claims 1-8, characterized in that: The system comprises a fouling composition structure feature acquisition module, a fouling composition structure feature analysis module, a descaling scheme formulation output module, and a descaling process online control feedback module. The dirt composition structure feature acquisition module comprises a dirt cross-section composition structure detection unit and a dirt cross-section composition structure feature image generation unit; the dirt cross-section composition structure detection unit acquires dirt composition structure detection image data of an axial cross-section inside the evaporator tube by using an X-ray detector; the dirt cross-section composition structure feature image generation unit generates dirt cross-section composition structure feature image data by using data smoothing processing and noise reduction on the dirt composition structure detection image data; The dirt composition structure feature analysis module comprises a dirt composition structure type feature image storage unit and a dirt composition structure type feature recognition output unit; the dirt composition structure type feature image storage unit is used for storing dirt composition structure type image data; the dirt composition structure type feature recognition output unit acquires dirt cross-section composition structure feature image data and dirt composition structure type image data, and uses an algorithm model to recognize and output dirt composition structure type feature data of the evaporator tube in the axial direction to the inner diameter circumferential direction; The descaling scheme development output module comprises a descaling flow and mode matching unit, a descaling operation process time measurement unit and a descaling operation overall scheme output unit; the descaling flow and mode matching unit matches dirt composition structure type feature data to form a mode and a flow for descaling each dirt layer; the descaling operation process time measurement unit is used for measuring an average thickness of each dirt layer in the dirt cross-section composition structure feature image data, and measuring a descaling operation time according to the average thickness; the descaling operation overall scheme output unit is used for integrating dirt composition structure type feature data, a descaling mode and flow and a descaling time to form and output a descaling operation scheme; The descaling process online control feedback module comprises a post-descaling dirt composition structure feature existence recognition unit and a descaling result detection judgment unit; the post-descaling dirt composition structure feature existence recognition unit is used for judging and recognizing a measured average thickness of a dirt layer according to real-time acquired dirt cross-section composition structure feature image data after descaling operation; the descaling result detection judgment unit performs online feedback output on a descaling operation result.

Images