A heat exchanger descaling and anti-corrosion system and device based on intelligent control

Abstract

The present invention discloses a heat exchanger descaling and anti-corrosion system and device based on intelligent control, which relates to the field of system control technology. The heat exchanger descaling and anti-corrosion system based on intelligent control monitors the key parameters of the heat exchanger through a real-time monitoring module to effectively prevent failures and performance degradation; the abnormality recognition module detects abnormalities based on key parameters, and builds a model through historical abnormal image data to determine the real-time abnormal type and scaling thickness, so as to achieve rapid fault identification; the control module dynamically adjusts the pulse width and period of the power oscillation wave according to the fluid viscosity data to automatically descale and prevent corrosion of the heat exchanger; the corrosion prediction module predicts the corrosion rate and time based on the anti-corrosion module data, provides a reference for maintenance, optimizes equipment management, reduces maintenance costs, and ensures the continuity and safety of the system. The system integrates real-time monitoring, intelligent identification, automatic anti-corrosion and descaling, and predictive control, which improves the operating efficiency and reliability of the device.

Description

Technical Field

[0001] The present invention relates to the technical field of system control, and in particular to a heat exchanger descaling and anti-corrosion system and device based on intelligent control. Background Art

[0002] In real time, a heat exchanger is a device used to transfer heat between two or more fluids. It is widely used in many fields of industry, commerce and home. Its main purpose is to transfer heat from hot fluid to cold fluid to achieve temperature regulation; reduce equipment energy consumption and improve system efficiency through efficient heat exchange; long-term use of heat exchangers will cause scaling;

[0003] Traditional scaling treatment often has difficulty in dealing with abnormal data of particularly hard or strong adhesion. Using chemical cleaning treatment not only damages the equipment, but also pollutes the environment. At the same time, heat leakage and explosion caused by corrosion of the heat exchanger also occur from time to time, causing heavy losses to people's lives and property, causing wear or damage to the scaling parts of the heat exchanger, affecting its lifespan;

[0004] The present invention automatically adjusts the pulse width and period of the power oscillation wave according to the fluid viscosity data to automatically remove scale and prevent corrosion of the heat exchanger. At the same time, the corrosion efficiency and corrosion time of the heat exchanger are predicted to ensure that the heat exchanger can be maintained in time to avoid affecting the operation of the equipment.

[0005] In summary, how to automatically perform corrosion prevention and descaling of heat exchangers according to fluid viscosity and predict corrosion efficiency and time to ensure the normal operation of the equipment is an urgent problem to be solved and optimized for heat exchanger descaling and anti-corrosion devices based on intelligent control. Summary of the invention

[0006] The present invention provides a heat exchanger descaling and anti-corrosion system and device based on intelligent control, which solves the technical problem of how to automatically perform anti-corrosion and descaling on the heat exchanger according to the fluid viscosity and predict the corrosion efficiency and time to ensure the normal operation of the equipment.

[0007] In order to solve the above technical problems, the present invention provides a heat exchanger descaling and anti-corrosion system and device based on intelligent control. The specific technical solution is as follows:

[0008] In a first aspect, a heat exchanger descaling and anti-corrosion system based on intelligent control comprises:

[0009] Real-time monitoring module 201: used to monitor the key parameters of the heat exchanger in real time; the key parameters include the temperature, flow rate and water quality component parameters of the heat exchanger;

[0010] Abnormal identification module 202: It is used to collect abnormal image data based on the abnormal key parameters of the heat exchanger monitored by the real-time monitoring module; construct a training set with the abnormal image data of the heat exchanger history to obtain an abnormal identification model to output abnormal type data; input the abnormal image data collected in real time into the abnormal identification model to output the identification result of the real-time heat exchanger abnormal type data; and judge the scaling thickness based on the identification result;

[0011] Control module 203: It is used to obtain real-time scaling thickness information data according to the real-time abnormal type data identification result of the heat exchanger, and automatically perform scaling and corrosion prevention treatment on the heat exchanger by self-adjusting the pulse width and period of the power oscillation wave according to the real-time acquired fluid viscosity data, so as to prevent scaling adhesion; and eliminate stress corrosion of the heat exchanger;

[0012] The corrosion prediction module 204 is used to adjust the power oscillation wave pulse width and period results based on the control module to predict the corrosion rate and corrosion time of the heat exchanger to prevent the heat exchanger from being corroded and affecting the operation of the equipment.

[0013] As a further optimization solution of the present invention, the real-time monitoring module further includes:

[0014] Monitoring unit, collection unit and pre-processing unit;

[0015] The monitoring unit monitors the temperature, flow rate and water quality component parameter data of the heat exchanger in real time, sets the standard threshold of each parameter, and when the monitored parameter data is greater than the standard threshold, it indicates that the parameter data is abnormal;

[0016] Based on the monitored parameter abnormal data, the acquisition unit acquires the parameter abnormal image data of the heat exchanger to obtain the abnormal image data;

[0017] Based on the abnormal image data, the preprocessing unit preprocesses the abnormal image data to obtain clear abnormal image data.

[0018] As a further optimization solution of the present invention, the abnormality identification module further includes:

[0019] Analysis unit, recognition unit and judgment unit;

[0020] The analyzing unit extracts features from the acquired clear abnormal image data to obtain feature image data;

[0021] Inputting the characteristic image data of the heat exchanger history into the recognition unit to train and construct the abnormality recognition model; the abnormality recognition model outputs the recognition result of the abnormality type data of the heat exchanger;

[0022] Based on the identification result of the abnormal type data, the judgment unit judges the scaling thickness information data through the scaling type characteristic data.

[0023] As a further optimization solution of the present invention, based on the identification result of the abnormal type data, the judgment unit obtains scaling type characteristic data to judge scaling thickness information data, including:

[0024] According to the recognition result, the judgment unit obtains the characteristic data of the scaling area, wherein the characteristic data includes the shape, color and distribution area data of the scaling;

[0025] Based on the determination unit, the scaling area characteristic data is obtained, and the scaling thickness information data is obtained by extracting the RGB color mode data of the scaling area, analyzing the color chromaticity C, brightness L and hue θ parameters of the scaling area, and combining the scaling density information data D. ; Where f is the conversion function.

[0026] As a further optimization solution of the present invention, the control module further includes:

[0027] Based on the real-time scaling thickness information data, heat exchange efficiency data of the heat exchanger is obtained;

[0028] Based on the scaling thickness information data and the heat exchange efficiency data of the heat exchanger, the real-time fluid viscosity data is obtained. Deviation analysis is performed to obtain the impact data of the deviation between fluid viscosity and heat exchange demand of the heat exchanger; where μ represents the real-time fluid viscosity; SR(t) is the time viscosity reading of the viscosity sensor at the tth moment;

[0029] Based on the impact data, a preset threshold is set, and the impact data is compared with the preset threshold; based on the comparison result, the control module self-adjusts the pulse width and period of the power oscillation wave to obtain an automatic descaling and anti-corrosion strategy for the heat exchanger.

[0030] As a further optimization scheme of the present invention, based on the influence data, the control module self-adjusts the pulse width and period of the power oscillation wave to obtain an automatic descaling and anti-corrosion strategy for the heat exchanger, and further includes:

[0031] According to the comparison result of the influence degree data, when the influence degree data is greater than or equal to the preset threshold, the real-time fluid viscosity data μ is too high. By increasing the power shock wave intensity by more than 30%, the wavelength is increased to impact the scaling layer with tight adhesion.

[0032] Based on increasing the power shock wave intensity, through: ; Increase the pulse width to 300μs so that the power shock wave can further impact the scale until it falls off; and automatically increase the disturbance effect of the power shock wave to reduce the contact time between the fluid corrosive medium and the metal surface;

[0033] Where, w represents the real-time pulse width of the oscillating wave; w0 represents the initial pulse width; K w is the pulse width adjustment coefficient, which indicates the sensitivity of pulse width to viscosity change; μ opt Indicates optimum or target viscosity;

[0034] When the impact data is less than the preset threshold, the real-time fluid viscosity data μ is too low, through: ; Increase the frequency of the power oscillation wave by 20-40Hz; and reduce the pulse width by 100μs to allow the fluid to circulate more quickly and reduce the retention of corrosive media;

[0035] In the formula, g represents the real-time oscillation frequency; g0 represents the initial oscillation frequency; K g Indicates the frequency adjustment coefficient, which indicates the sensitivity of frequency to viscosity changes; μ opt Indicates optimum or target viscosity;

[0036] According to the viscosity change through: ; To adjust the control parameters of the power oscillation wave so that the control module tends to a stable state;

[0037] Where, u(t) is the control signal at time t, which represents the control signal for adjusting the frequency or pulse width of the oscillation wave; K p Represents proportional gain; K i represents differential gain; K d represents the interaction gain; represents the cumulative deviation; Indicates the rate of change of the deviation.

[0038] As a further optimization solution of the present invention, the corrosion prediction module further includes:

[0039] Acquire historical viscosity data and train to build a trend prediction model; collect real-time viscosity data of the fluid; input the real-time viscosity data into the trend prediction model to output the recognition result of the corrosion degree of the heat exchanger based on the viscosity data; predict the corrosion trend of the heat exchanger based on the recognition result;

[0040] Based on the trend prediction model, the changing trend of the corrosion effect of the viscosity data on the heat exchanger (3) is predicted to quantify the corrosion efficiency and corrosion time node of the fluid viscosity on the heat exchanger; and based on the corrosion time node, the heat exchanger is maintained in advance.

[0041] As a further optimization scheme of the present invention, historical viscosity data is obtained and trained to construct a trend prediction model; real-time viscosity data of the fluid is collected; the real-time viscosity data is input into the trend prediction model to output an identification result of the corrosion degree of the heat exchanger by the viscosity data; based on the identification result, the corrosion trend of the heat exchanger is predicted, including:

[0042] The historical viscosity data is encoded into sequence data, and the sequence data is input into the trend prediction model; the trend prediction model includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the intermediate representation data of the hidden layer is transmitted to the output layer, and the output layer output represents the corrosion degree recognition result of the viscosity data, which is as follows:

[0043] , where W u , W r , W c represents the weight parameter, b u 、b r 、b c represents the bias parameter, represents the dot product, u （t） 、r （t） and c （t） Represent the intermediate states of the first, second, and third hidden layers respectively, X （t） The data item representing the viscosity data at the tth moment, H （t） and H （t-1） They represent the intermediate representation data of the viscosity data at the t-th and t-1-th moments, n ≥ t ≥ 1, n represents the total number of training input viscosity data, and H （t-1） =X （t） , tanh is the hyperbolic tangent function, σ represents the sigmoid function;

[0044] Based on the constructed trend prediction model, the acquired real-time viscosity data is input into the trend prediction model to output the corrosion degree identification result of the real-time viscosity data;

[0045] The corrosion degree identification results of real-time viscosity data are compared with those of historical viscosity data to determine the development trend of the corrosion degree of the heat exchanger due to viscosity changes.

[0046] As a further optimization scheme of the present invention, the trend of the influence of viscosity data on the corrosion of the heat exchanger (3) is predicted based on the trend prediction model, so as to quantify the corrosion efficiency and corrosion time node of the fluid viscosity on the heat exchanger; and the heat exchanger is maintained in advance according to the corrosion time node, including:

[0047] Based on the development trend, according to the real-time fluid viscosity, fluid composition, fluid velocity, pulse width and power shock wave data, through: ; to obtain the corrosion rate; R represents the corrosion rate; k represents the corrosion rate constant; E q represents the fluid flow rate; T represents the temperature; F represents the ideal fluid component category parameter set; e represents the power shock wave data; R T Indicates corrosion temperature;

[0048] When the allowable corrosion depth of the heat exchanger is d, the corrosion time node is estimated according to the corrosion rate, by: ; To obtain the corrosion time node t;

[0049] When the estimated corrosion time node t is lower than the preset safety time, the trend prediction module issues an alarm.

[0050] In the second aspect, a heat exchanger descaling and anti-corrosion device based on intelligent control is provided. The high-intensity power wave generated by the device can eliminate the laminar layer in the heat exchanger, convert heat conduction into convection heat transfer, and improve the heat exchange efficiency by more than 10%. The device is used to implement a heat exchanger descaling and anti-corrosion system based on intelligent control, and the device specifically includes:

[0051] Intelligent digital electric control cabinet, energy conversion probe, heat exchanger, link wire, digital instrument wire and digital intelligent display screen;

[0052] One end of the link wire is connected to the intelligent digital electric control cabinet, and the other end is connected to the energy conversion probe; the intelligent digital electric control cabinet sends a telecommunication signal which is transmitted to the energy conversion probe through the link wire, so that the telecommunication signal is converted into a power oscillation wave; the power oscillation wave automatically removes scale and prevents corrosion on the heat exchanger;

[0053] The energy conversion probe is arranged on the heat exchanger and is tightly connected with the heat exchanger; a plurality of plates and a plurality of tubes are arranged in the heat exchanger, and the tubes are arranged through the plates. The energy conversion probe transmits the power shock wave to each plate and each tube. The power shock wave runs at a high speed in the plates and tubes so that the particles in the plates collide with each other to eliminate stress corrosion; and micro-impact is generated on the inside and outside of the plates and tubes to prevent scaling from approaching and adhering to the heat exchanger;

[0054] The intelligent digital electric control cabinet is provided with a viscosity meter, which is connected to the heat exchanger through a digital instrument wire to detect the viscosity of the fluid in the heat exchanger; the digital intelligent display screen is provided on the intelligent digital electric control cabinet to display the fluid viscosity data;

[0055] The fluid viscosity data is transmitted to the remote mobile phone control terminal through the chip, and the remote mobile phone control terminal remotely adjusts the pulse width and period of multiple gears of the power oscillation wave corresponding to the fluid viscosity to achieve the best descaling and anti-corrosion effect.

[0056] The present invention has at least the following beneficial effects: the present invention monitors the key parameters of the heat exchanger in real time through the real-time monitoring module; the key parameters include the temperature, flow rate and water quality component parameters of the heat exchanger; thereby effectively preventing failure or performance degradation. Real-time monitoring not only improves the response speed of the system, but also provides more accurate working condition data to support subsequent judgment and processing.

[0057] The abnormality recognition module collects abnormal image data based on the abnormal key parameters of the heat exchanger monitored by the real-time monitoring module; constructs a training set with the abnormal image data of the heat exchanger history to obtain an abnormality recognition model to output abnormal type data; inputs the abnormal image data collected in real time into the abnormality recognition model to output the recognition result of the real-time heat exchanger abnormal type data; judges the scaling thickness based on the recognition result; thus quickly judges the nature of the fault, especially can identify and quantify the scaling thickness. In this way, the system can respond immediately when problems such as scaling are detected, avoiding the inconvenience and delay of manual inspection.

[0058] The control module obtains real-time scaling thickness information data based on the real-time abnormal type data identification results of the heat exchanger. Based on the real-time fluid viscosity data, the heat exchanger is automatically descaled and anti-corroded by self-adjusting the pulse width and period of the power oscillation wave to prevent scaling adhesion and eliminate stress corrosion of the heat exchanger. This module helps prevent scaling from adhering again and effectively eliminates the impact of stress corrosion on the equipment, extending the service life of the heat exchanger. The control module dynamically adjusts the fluid viscosity during operation to ensure the continuity and stability of the cleaning and anti-corrosion effects.

[0059] The corrosion prediction module adjusts the power oscillation pulse width and cycle results based on the control module to predict the corrosion rate and corrosion time of the heat exchanger to avoid the corrosion of the heat exchanger affecting the operation of the equipment; it provides a reference for maintenance in advance to reduce the interference of corrosion on the normal operation of the equipment. This prediction function can achieve forward-looking equipment management, thereby optimizing maintenance frequency and planning, reducing maintenance costs, and improving the continuity and safety of equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of a flow chart of a heat exchanger descaling and anti-corrosion system based on intelligent control provided by an embodiment of the present invention;

[0061] Figure 2 It is a schematic diagram of a heat exchanger descaling and anti-corrosion device based on intelligent control provided by an embodiment of the present invention;

[0062] Figure 3 It is a flow chart of a heat exchanger descaling and anti-corrosion system method based on intelligent control provided by an embodiment of the present invention.

[0063] Attached photos

[0064] 1. Intelligent digital electric control cabinet; 2. Energy conversion probe; 3. Heat exchanger; 4. Link wire; 5. Digital instrument wire; 6. Digital intelligent display screen. DETAILED DESCRIPTION

[0065] The present application is further described in detail below in conjunction with the accompanying drawings. It is necessary to point out here that the following specific implementation methods are only used to further illustrate the present application and cannot be understood as limiting the scope of protection of the present application. Technical personnel in this field can make some non-essential improvements and adjustments to the present application based on the above application content.

[0066] This embodiment provides a heat exchanger descaling and anti-corrosion system and device based on intelligent control, and the specific implementation methods are as follows:

[0067] like Figure 1 and Figure 3 As shown, a heat exchanger descaling and anti-corrosion system based on intelligent control includes:

[0068] Real-time monitoring module: it is used to monitor the key parameters of the heat exchanger 3 in real time; the key parameters include the temperature, flow rate and water quality component parameters of the heat exchanger 3;

[0069] Abnormal identification module: it is used to collect abnormal image data based on the abnormal key parameters of the heat exchanger 3 monitored by the real-time monitoring module; construct a training set with the abnormal image data of the heat exchanger 3 to obtain an abnormal identification model to output abnormal type data; input the abnormal image data collected in real time into the abnormal identification model to output the identification result of the abnormal type data of the heat exchanger 3 in real time; and judge the scaling thickness based on the identification result;

[0070] Control module: It is used to obtain real-time scaling thickness information data according to the real-time abnormal type data identification result of the heat exchanger 3, and automatically remove scale and prevent corrosion of the heat exchanger 3 by self-adjusting the pulse width and period of the power oscillation wave according to the real-time acquired fluid viscosity data, so as to prevent scaling adhesion; and eliminate stress corrosion of the heat exchanger 3;

[0071] Corrosion prediction module: It is used to adjust the power oscillation wave pulse width and period results based on the control module to predict the corrosion rate and corrosion time of the heat exchanger 3 to avoid the corrosion of the heat exchanger 3 affecting the operation of the equipment.

[0072] Specifically, the system cooperates with the device to effectively prevent the approach and adhesion of scale, so that the wall surface of the heat exchanger 3 is cleaned, the occurrence of under-scale corrosion is eliminated, and the power shock wave causes the particles in the plate to collide with each other, thereby eliminating stress corrosion.

[0073] By conducting online digital detection of fluid viscosity in the heat exchanger 3, the data is transmitted to a remote mobile phone through a chip. Based on the provided data (viscosity is set to five gears: low viscosity, medium viscosity, medium viscosity, high viscosity, and high viscosity), the pulse width and period of the five gears corresponding to the power wave are remotely controlled to achieve the best descaling and anti-corrosion effect of the device.

[0074] In this embodiment, the key operating parameters of the heat exchanger 3 are continuously monitored by using a sensor network through a real-time monitoring module. The key operating parameters include but are not limited to temperature, flow rate, and water quality composition. Monitoring of water quality composition is particularly important. It can provide information about mineral content, pH, etc., which directly affect the occurrence of scaling and corrosion rate. Through continuous parameter monitoring, the system can capture possible performance anomalies in the heat exchanger 3 at an early stage to ensure operating efficiency. The stability of temperature and pressure can help control the heat transfer efficiency, changes in flow rate can indicate the risk of blockage or leakage, and real-time data on water quality composition supports the subsequent modules to effectively deal with scaling and corrosion. The existence of this module enables the entire system to achieve a higher response speed and reduce downtime caused by equipment problems.

[0075] The anomaly recognition module receives data provided by the real-time monitoring module and immediately collects corresponding image data when an anomaly is detected. Subsequently, the system uses the historical anomaly image data collected in the past to build a deep learning model for identifying different types of anomalies. After inputting the real-time anomaly image data into the model, the system can identify specific anomaly types, such as scaling or corrosion, and judge the thickness of scaling based on the recognition results; through automated anomaly recognition and quantification of scaling thickness, the system can determine the cause of the fault more quickly and accurately. The recognition module enables problems such as scaling to be discovered and located in a timely manner, avoiding delays in manual inspection. At the same time, it provides the necessary scaling thickness information for subsequent modules to implement targeted cleaning and anti-corrosion operations. This data-driven anomaly recognition greatly improves the reliability of the system.

[0076] The control module, after the abnormal identification module provides information on the thickness of the scale and the degree of corrosion, the control module handles the scaling problem by adjusting the viscosity of the fluid. This operation can reduce the deposition of minerals, prevent the re-adhesion of scale, and at the same time reduce the risk of stress corrosion. Methods for adjusting the viscosity of the fluid may include using special descaling agents or adjusting the water flow rate; the control module can effectively prevent the heat exchanger 3 from failing due to scaling and reduce the structural damage caused by corrosion, thereby extending the life of the equipment. Compared with traditional regular cleaning and maintenance, this intelligent automatic descaling and anti-corrosion treatment saves costs while avoiding frequent shutdowns. By reducing the decrease in heat exchange efficiency caused by scaling and corrosion, the module ensures the continued efficient operation of the system.

[0077] After completing the automatic descaling and anti-corrosion operation, the corrosion prediction module uses the data from the cleaning process to estimate the future corrosion rate and scaling trend. It combines the real-time fluid viscosity adjustment results and historical data to establish a prediction model to determine the future state of the heat exchanger 3. This model can predict the corrosion rate, scaling cycle, etc., and then make an estimate of the remaining life of the heat exchanger 3. By predicting the possible corrosion and scaling development trends in advance, the system can formulate a more reasonable maintenance plan and reduce the risk of sudden failures. The forward-looking decision-making function of the corrosion prediction module helps equipment managers optimize maintenance cycles, reduce maintenance costs, and extend the service life of the equipment. By avoiding serious corrosion and scaling problems, the entire system can maintain a stable and efficient operating state.

[0078] Through the collaborative work between the above steps, the heat exchanger 3 management system can achieve all-round control from real-time monitoring, abnormality identification, active cleaning to predictive regulation. This multi-level design significantly improves the operating efficiency of the heat exchanger 3 and reduces maintenance and downtime costs. The overall system not only improves the reliability of the equipment, but also realizes active maintenance and preventive management, making the operation of the equipment safer and longer-lasting. It not only improves the operating efficiency and safety of the heat exchanger 3, but also realizes the functions of predictive and active maintenance, effectively extending the life of the equipment and reducing maintenance costs.

[0079] In a preferred embodiment of the present invention, the real-time monitoring module further includes:

[0080] Monitoring unit, collection unit and pre-processing unit;

[0081] The monitoring unit monitors the temperature, flow rate and water quality component parameters of the heat exchanger 3 in real time, sets the standard threshold of each parameter, and when the monitored parameter data is greater than the standard threshold, it indicates that the parameter data is abnormal; Specifically, the temperature, flow rate and water quality component parameters of the heat exchanger 3 are transmitted and circulated through the digital instrument wire 5;

[0082] Based on the monitored parameter abnormal data, the acquisition unit acquires the parameter abnormal image data of the heat exchanger 3 to obtain the abnormal image data;

[0083] Based on the abnormal image data, the preprocessing unit preprocesses the abnormal image data to obtain clear abnormal image data.

[0084] In the embodiment of the present invention, the monitoring unit tracks the temperature, flow rate and water quality components (such as pH, mineral content, etc.) of the heat exchanger 3 in real time through a series of sensors, and sets the standard threshold of temperature between 60°C and 100°C; the standard threshold of pressure between 7bar and 10bar; the standard threshold of water quality between 6.5 and 8.5; when a parameter exceeds the standard threshold, the unit identifies the data as abnormal data; by capturing changes at the moment when the parameter abnormality occurs, the monitoring unit can issue an early warning before the performance of the heat exchanger 3 deteriorates or potential damage occurs. This real-time monitoring not only reduces the inspection cycle and labor costs, but also helps avoid major failures and improves the overall safety and efficiency of the system.

[0085] When the monitoring unit finds that the parameters are abnormal, the acquisition unit starts to operate and collects abnormal image data of the heat exchanger 3. The image data can intuitively display the characteristics of the abnormal area (such as scaling, corrosion, etc.), providing a basis for subsequent data analysis and processing; the acquisition unit ensures that the image information of the real situation can be collected in time when the parameters are abnormal, which is helpful for further identification and processing. Compared with simple numerical values, image data is more intuitive, can assist in accurate diagnosis, reduce misjudgment, and provide detailed basic data for subsequent image recognition.

[0086] The preprocessing unit cleans and optimizes the abnormal image data obtained by the acquisition unit, such as removing noise and enhancing edges, to ensure the clarity of the image. The preprocessed image has higher quality and stronger recognition, which is conducive to subsequent algorithm analysis; the preprocessing unit greatly improves the quality of abnormal image data, making the subsequent abnormal type identification more accurate and reliable. High-quality images help reduce the error of the algorithm during recognition and reduce interference in the data analysis process, thereby ensuring that the system has high accuracy and stability when dealing with abnormal problems.

[0087] Through the division of labor and cooperation among the monitoring unit, the acquisition unit, and the pre-processing unit, all-round monitoring of the heat exchanger 3 and high-quality abnormal data collection can be achieved. This not only ensures the timely capture of abnormal situations, but also provides a clear and reliable data basis for subsequent intelligent identification and processing. The system as a whole improves the fault response speed, reduces the potential downtime risk of the heat exchanger 3, and ensures the safety and efficiency of operation.

[0088] In a preferred embodiment of the present invention, the abnormality identification module further includes:

[0089] Analysis unit, recognition unit and judgment unit;

[0090] The analyzing unit extracts features from the acquired clear abnormal image data to obtain feature image data;

[0091] Inputting the characteristic image data of the heat exchanger 3 into the recognition unit to train and construct the abnormality recognition model; the abnormality recognition model outputs the recognition result of the abnormality type data of the heat exchanger 3;

[0092] Based on the identification result of the abnormal type data, the judgment unit judges the scaling thickness information data through the scaling type characteristic data.

[0093] In an embodiment of the present invention, an analysis unit receives clear abnormal image data obtained from a sensor or a camera device; then uses image processing technology (such as edge detection, texture analysis, color analysis, etc.) to analyze the abnormal image and extract important features to capture features of different scales and complexities; significance and effect; provides accurate feature data for subsequent recognition to ensure that the details of the abnormal area in the image are fully captured; feature extraction improves the availability of data, so that subsequent recognition units can perform model training and abnormality recognition based on accurate data.

[0094] The recognition unit inputs the historical characteristic image data of the heat exchanger 3 into the model for training, and constructs and optimizes the abnormal recognition model; in real-time applications, the trained model is used to analyze the newly input image and output the recognition result of the abnormal type data; the abnormal type (scaling or corrosion) in the image is quickly identified, and efficiency is improved through automated processing; accurate classification of abnormalities helps to provide early warning and adopt corresponding maintenance strategies, thereby extending the life of the equipment.

[0095] The judgment unit receives the abnormal type data output by the identification unit, and evaluates the scaling thickness information by comparing it with the scaling type characteristic data to determine the thickness level of different types of scaling; it provides detailed abnormal analysis results, which not only identifies the abnormal type but also quantifies its severity (such as scaling thickness); it helps to formulate accurate maintenance and cleaning plans and optimize the operation and maintenance strategy of the heat exchanger 3.

[0096] Through the integration of the three units, the system can effectively detect and identify abnormal conditions in the heat exchanger 3 in real time. Through feature extraction of the analysis unit, abnormal classification of the identification unit and thickness analysis of the judgment unit, the entire system can achieve high-accuracy abnormality detection through precise feature extraction and model recognition; identify potential problems in advance and correctly assess their severity to help formulate more targeted maintenance measures; and reduce unnecessary downtime and improve equipment operation efficiency through efficient identification and judgment of abnormalities.

[0097] The abnormality identification model specifically includes: generating two-dimensional structure data from the collected abnormality type data, the two-dimensional structure data includes a data matrix and a relationship matrix, one unit of the data matrix represents one-dimensional structure data of an independent object, the independent object includes a heat exchanger 3 and abnormality type data, one unit of the data matrix representing the heat exchanger 3 only includes the abnormality type data of the heat exchanger 3, and one unit of the data matrix representing the abnormality type data only includes the abnormality identification information of the abnormality type data;

[0098] The element in the i-th row and j-th column of the relationship matrix represents the association between the i-th unit of the data matrix and the independent object represented by the j-th unit. If there is an association, the value of this element of the relationship matrix is ​​1, otherwise it is 0;

[0099] The association between heat exchanger 3 and abnormal type data means that heat exchange efficiency of heat exchanger 3 will be affected by abnormal type data, as follows:

[0100] , where K v represents the intermediate representation data of the vth unit of the data matrix, M （v） The set of cells of the data matrix that are associated with the vth cell of the data matrix, x u The data item representing the abnormal type data represented by the u-th unit of the data matrix, C v is a normalization constant (such as the unit degree of the data matrix), b represents the weight parameter, and σ is the sigmoid function.

[0101] In a preferred embodiment of the present invention, the above-mentioned determination unit further includes:

[0102] According to the recognition result, the judgment unit obtains the characteristic data of the scaling area, wherein the characteristic data includes the shape, color and distribution area data of the scaling;

[0103] Based on the determination unit, the characteristic data of the scaling area is obtained, and the RGB color mode data of the scaling area is extracted, and the color chromaticity C, brightness L and hue θ parameters of the scaling area are analyzed to calculate the position of the color on the color wheel, and the scaling density information data D is combined to obtain the scaling thickness information data. .

[0104] In this embodiment, after the judgment unit identifies the abnormality according to the recognition unit, it starts to collect characteristic data of the scaling area, including the shape, color and distribution area of ​​the scaling. These data reflect the basic form and coverage area of ​​the scaling, and provide a basis for further thickness judgment; through the data analysis of shape, color and area, the system can intuitively evaluate the distribution of scaling and provide a clear characteristic basis for subsequent color and thickness calculation.

[0105] By analyzing the RGB color mode data of the scaling area, the system obtains the red, green and blue color values ​​of the scaling area. Different color combinations can reflect the composition and changes of scaling; RGB data provides a detailed division of scaling colors, which helps the system to further distinguish scaling types, especially in different lighting and environments. It can still maintain a high recognition accuracy.

[0106] Analyze color chromaticity (C): ; thus reflecting the proportion of each color component in the total color; where R, G, B are color channel values, ranging from 0 to 255;

[0107] Brightness (L): ; to account for the different sensitivities and tones of red, green, and blue in the human eye;

[0108] Hue (θ) parameter: ; To calculate the position of the color on the color wheel;

[0109] Chroma (C) reflects the vividness of the color, brightness (L) represents the degree of lightness and darkness, and hue (θ) indicates the position of the color on the color wheel. After analyzing these three parameters, the system can more accurately understand the color characteristics of the scale; this refined color parameter analysis helps the system to more accurately distinguish the nature of the scale. For example, some scales have characteristic differences in chromaticity and brightness, which can be used as a basis for determining the composition of the scale.

[0110] Based on the hue (θ) parameter, the system can accurately determine the position of the color in the color wheel. This step of analysis can compare the color of the scale with the standard color wheel to obtain the unique color position of the scale; by comparing the color wheel position, the system can quickly match the special color area of ​​the scale area, and then infer the chemical properties or composition of the scale, improving the accuracy of identifying the type of scale.

[0111] Combined with scaling density information data (D): Where, P scale is the number of pixels in the fouling area, P total is the total number of pixels in the area to obtain scaling thickness information; the system combines color parameters with density information to comprehensively judge scaling thickness. Scaling density data (D) indicates the density of scaling per unit area. When the density is high, it usually means that the scaling layer is thicker; after combining density data, the system can more accurately estimate the scaling thickness and use the results for real-time maintenance of equipment. Thickness information is crucial for timely cleaning of scaling and avoiding reduced equipment efficiency.

[0112] In a preferred embodiment of the present invention, the control module further includes:

[0113] Based on the real-time scaling thickness information data, heat exchange efficiency data of the heat exchanger 3 is obtained;

[0114] Based on the scaling thickness information data and the heat exchange efficiency data of the heat exchanger 3, the real-time fluid viscosity data Deviation analysis is performed to obtain the influence data of the deviation between fluid viscosity and the heat exchange demand of heat exchanger 3; wherein μ represents the real-time fluid viscosity; SR(t) is the time viscosity reading of the viscosity sensor at the tth moment;

[0115] Based on the impact data, a preset threshold is set, and the impact data is compared with the preset threshold; based on the comparison result, the control module self-adjusts the pulse width and period of the power oscillation wave to obtain an automatic descaling and anti-corrosion strategy for the heat exchanger 3.

[0116] In this embodiment, the presence of scaling will affect the heat exchange efficiency of the heat exchanger 3 by: ; to obtain the thermal resistance coefficient of the structure; where R 结垢 represents scaling thermal resistance; k 结垢 represents the thermal conductivity of scaling; A represents the scaling area; Indicates the scaling thickness;

[0117] Then through: ; R 总 represents the thermal resistance of the real-time heat exchanger 3; R 初始 Represents the total thermal resistance when there is no scaling; finally by: ; To obtain the real-time actual heat exchange efficiency coefficient; The heat exchange efficiency of the heat exchanger 3 is calculated through the scaling thickness information; A thicker scaling layer will hinder heat transfer and reduce heat exchange efficiency, so there is a negative correlation between scaling thickness and heat exchange efficiency. According to the real-time heat exchange efficiency data; By real-time monitoring of scaling thickness and adjusting the heat exchange efficiency accordingly, it is possible to provide an accurate evaluation of heat exchange performance, thereby helping operators to take cleaning measures at the right time and improve system efficiency.

[0118] Fluid viscosity will affect the flow and heat exchange process of the fluid in the heat exchanger 3. Too high or too low viscosity will reduce the heat exchange efficiency. By analyzing the scaling thickness information and real-time heat exchange efficiency data, using: ; In the formula, ∆Q represents the deviation of heat exchange demand; ∆μ represents the change of viscosity; thus, the deviation effect of fluid viscosity on heat exchange demand can be inferred. Real-time fluid viscosity data is compared with heat exchange demand to evaluate the effect of fluid viscosity on the heat exchange process; deviation analysis helps the system better understand the effect of fluid characteristics on the performance of heat exchanger 3, provides accurate adjustment basis, and ensures that heat exchanger 3 operates under optimal conditions.

[0119] Through the above deviation analysis, the system generates deviation impact data of fluid viscosity and heat exchange demand of heat exchanger 3. This is a key indicator obtained by analyzing the difference between fluid viscosity and heat exchange demand, describing the actual impact of real-time viscosity on the performance of heat exchanger 3, including:

[0120] When the fluid viscosity increases, the fluidity of the fluid decreases, causing the flow pattern in the heat exchanger 3 to be laminar and the mixing effect to be poor. As a result, the velocity gradient of the fluid near the wall of the heat exchanger 3 decreases, resulting in a decrease in the convective heat transfer coefficient, affecting the heat exchange efficiency.

[0121] Reynolds number change: The heat transfer coefficient is positively correlated with the Reynolds number, and the Reynolds number is affected by the fluid viscosity. ; Where ρ is the fluid density; v is the flow velocity; D is the characteristic length of the heat exchanger 3; When μ viscosity increases, the Reynolds number decreases, affecting the heat transfer coefficient and thus the heat transfer efficiency. When the viscosity increases, the Reynolds number decreases, causing the flow pattern to change from turbulent to laminar, further reducing the heat transfer efficiency.

[0122] Higher viscosity increases the resistance to flow of the fluid in the heat exchanger 3, causing a greater pressure drop. This phenomenon is particularly evident in applications that require high flow rates, because higher flow resistance requires more pump power to maintain the flow rate, thereby increasing the operating cost of the system; to overcome the increased pressure drop, the pump power demand will also increase, resulting in higher energy consumption. For example, when the viscosity increases to twice its original value, the energy consumption of the pump may increase exponentially, which has a significant impact on the operating cost of the system.

[0123] If the heat exchanger 3 is designed based on a preset viscosity, but the actual viscosity deviates significantly, the heat exchanger 3 may not operate under optimal conditions. High-viscosity fluids that were not considered during design may result in insufficient heat exchange or excessive pressure drop, affecting the overall performance of the heat exchanger 3; sometimes, after the viscosity increases, the flow rate needs to be adjusted to achieve better heat exchange. For example, the turbulent state can be increased by increasing the flow rate, thereby partially compensating for the decrease in heat transfer efficiency. However, this adjustment may be limited by physical conditions or energy consumption in some systems.

[0124] High-viscosity fluids tend to form boundary layers more easily, resulting in low flow velocity and high temperature gradient on the surface of the heat exchanger 3, which easily forms a scaling layer. Scaling will further increase the heat transfer resistance, reduce the efficiency of the heat exchanger 3, and increase the frequency of cleaning and maintenance; due to the impact of high-viscosity fluids on the heat transfer surface, the heat exchanger 3 may require more frequent cleaning and maintenance. The scaling layer not only affects the heat transfer efficiency, but also increases the overall operating cost of the equipment and reduces economic efficiency.

[0125] Case: In the oil industry, crude oil has a high viscosity, especially at low temperatures. This high viscosity affects the heat transfer efficiency, so efficient turbulence promotion technology needs to be used when designing heat exchanger 3. On the contrary, if the viscosity is too low, it may cause excessive flow rate and vibration, which in turn affects the stability of the equipment; in dairy processing, high-viscosity dairy products (such as cream) tend to form a thermal resistance layer when heated in heat exchanger 3, requiring a more complex heat exchanger 3 structure and a higher maintenance frequency to ensure production efficiency; the impact data quantifies the impact of viscosity on heat transfer efficiency, which facilitates the system to adjust and optimize according to specific conditions.

[0126] A threshold is preset, which is K•∆Q; where K is the deviation rate coefficient (5% to 10% fluid viscosity deviation rate or 3% to 5% heat exchange demand deviation rate); to determine whether the fluid viscosity deviation is within an acceptable range. By comparing the impact data with the threshold, the system can determine whether adjustments are needed. If the deviation is too large, it means that the fluid viscosity has significantly affected the working efficiency of the heat exchanger 3, and measures need to be taken; this automated threshold comparison mechanism can monitor the changes in fluid viscosity in real time, detect and handle problems in a timely manner, and avoid efficiency losses or equipment damage caused by viscosity changes.

[0127] Based on the results of the comparison between the impact data and the threshold, the system self-adjusts the pulse width and period of the power oscillation wave of the control module. The power oscillation wave is used for physical descaling. By adjusting the waveform parameters, the system can accurately control the descaling intensity, thereby protecting the equipment from excessive or insufficient oscillation. In addition, the adjustment of the fluid viscosity parameters can optimize the flow characteristics of the fluid and further improve the heat exchange efficiency; by automatically adjusting the descaling and anti-corrosion strategy, the system can achieve intelligent descaling, which can not only maintain the efficient operation of the heat exchanger 3, but also effectively avoid equipment corrosion or damage, and extend the service life of the heat exchanger 3.

[0128] In a preferred embodiment of the present invention, the above automatic descaling and anti-corrosion strategy further comprises the following steps:

[0129] According to the comparison result of the influence degree data, when the influence degree data is greater than or equal to the preset threshold, the real-time fluid viscosity data μ is too high. By increasing the power shock wave intensity by more than 30%, the wavelength is increased to impact the scaling layer with tight adhesion.

[0130] Based on increasing the power shock wave intensity, through: ; Increase the pulse width to 300μs so that the power shock wave can further impact the scale until it falls off; and automatically increase the disturbance effect of the power shock wave to reduce the contact time between the fluid corrosive medium and the metal surface;

[0131] Where, w represents the real-time pulse width of the oscillating wave; w0 represents the initial pulse width; K wis the pulse width adjustment coefficient, which indicates the sensitivity of pulse width to viscosity change; μ opt Indicates optimum or target viscosity;

[0132] When the impact data is less than the preset threshold, the real-time fluid viscosity data μ is too low, through: ; Increase the frequency of the power oscillation wave by 20-40Hz; and reduce the pulse width by 100μs to allow the fluid to circulate more quickly and reduce the retention of corrosive media;

[0133] In the formula, g represents the real-time oscillation frequency; g0 represents the initial oscillation frequency; K g Indicates the frequency adjustment coefficient, which indicates the sensitivity of frequency to viscosity changes; μ opt Indicates optimum or target viscosity;

[0134] According to the viscosity change through: ; To adjust the control parameters of the power oscillation wave so that the control module tends to a stable state;

[0135] Where, u(t) is the control signal at time t, which represents the control signal for adjusting the frequency or pulse width of the oscillation wave; K p Represents proportional gain; K i represents differential gain; K d represents the interaction gain; represents the cumulative deviation; Indicates the rate of change of the deviation.

[0136] In this embodiment, based on the comparison between the "influence data" (usually an indicator for measuring viscosity) and the preset threshold, it is determined whether the real-time viscosity data μ is too high or too low; the degree of deviation of the viscosity is accurately identified to provide a basis for subsequent oscillation wave adjustment, ensuring that appropriate parameter adjustment methods can be adopted.

[0137] When the viscosity is high, the intensity of the shock wave is increased by more than 30% to increase the wavelength and amplitude, thereby generating a greater impact force. The high-intensity shock wave can more effectively impact and destroy the scaling layer, especially for the tightly adhered scaling, with a higher removal efficiency, reduce the thickness of the scaling layer, enhance the thermal conductivity of the fluid, reduce the flow resistance to the heat exchanger 3 or the pipeline, and improve the heat transfer efficiency.

[0138] While increasing the intensity of the shock wave, the energy of the shock wave is more concentrated by increasing the pulse width; the pulse width refers to the duration of the waveform in an oscillation cycle. Increasing the pulse width will extend the action time of each shock wave, thereby increasing the impact on scaling; further strengthening the removal effect of the scaling layer, making the scaling fall off, reducing flow resistance, and reducing system energy consumption; by automatically increasing the disturbance effect of the shock wave, the contact time between the corrosive medium in the fluid and the metal surface is reduced;

[0139] By increasing the disturbance effect, the flow frequency of the fluid in the pipeline will be accelerated, making it difficult for corrosive media to stay on the metal surface for a long time, reducing the occurrence of corrosion; effectively extending the service life of the equipment, reducing the need for equipment maintenance caused by corrosion, and reducing the overall operating cost.

[0140] When the viscosity is low, the flow of the fluid is accelerated by increasing the frequency of the oscillation wave by 20-40Hz; the increase in frequency leads to an increase in the number of oscillations of the waveform, thereby increasing the kinetic energy of the fluid and making it flow faster; accelerating the flow of the fluid, reducing the residence time of the corrosive medium, reducing the risk of corrosion, and ensuring that the viscosity remains within a reasonable range, improving the heat transfer effect; while increasing the frequency, reduce the pulse width to allow the fluid to pass more quickly.

[0141] The reduction in pulse width means that the action time of each waveform is shortened. Combined with a higher frequency, the oscillation wave has a lighter and more frequent effect on the fluid, reducing the tendency to scale, while maintaining good fluid fluidity and preventing uneven heat transfer caused by low viscosity.

[0142] The control signal u(t) (the control signal at time t) is used to dynamically adjust the frequency and pulse width. The adjustment degree of the oscillation wave is controlled by proportional gain, differential gain, and interactive gain to achieve real-time feedback control of viscosity; ensure that the state of the anti-corrosion and descaling module tends to be stable, and through real-time monitoring and adjustment, the viscosity is kept near the target value to optimize the stability and efficiency of the system.

[0143] By adjusting the intensity, frequency and pulse width of the shock wave, we can flexibly respond to changes in scaling and fluid viscosity and improve the heat exchange efficiency of the equipment. By accelerating fluid flow and reducing the retention of corrosive media, we can reduce the risk of corrosion and extend the service life of the heat exchange equipment. Viscosity adjustment reduces fluid resistance and reduces pump power requirements, thereby saving energy. The automatic adjustment function of control parameters reduces manual intervention, realizes intelligent and automated management, and ensures efficient operation of the system.

[0144] In a preferred embodiment of the present invention, the corrosion prediction module further includes:

[0145] Obtain historical viscosity data and train to build a trend prediction model; collect real-time viscosity data of the fluid; input the real-time viscosity data into the trend prediction model to output the viscosity data to identify the corrosion degree of the heat exchanger 3; predict the corrosion trend of the heat exchanger 3 based on the identification result;

[0146] Based on the trend prediction model, the change trend of the corrosion effect of the viscosity data on the heat exchanger (3) is predicted to quantify the corrosion efficiency and corrosion time node of the fluid viscosity on the heat exchanger 3; according to the corrosion time node, the heat exchanger 3 is maintained in advance.

[0147] In this embodiment, a historical data set is formed by collecting and storing the viscosity data of the fluid at different time points. This data is used to train machine learning or statistical models to identify the trend of viscosity changes over time or working conditions; time series analysis methods are usually used to capture the law of viscosity changes; the possibility of future viscosity changes can be predicted by training trend models, providing data support for subsequent corrosion trend identification, and avoiding the impact of unexpected viscosity fluctuations on equipment performance.

[0148] Obtain the real-time viscosity value of the fluid, usually measured through online sensors or regular sampling to ensure the timeliness of the viscosity data; monitor the viscosity changes of the fluid in real time to ensure that data updates can reflect the actual working conditions, provide accurate input for trend prediction, and improve the reliability of prediction results.

[0149] The latest viscosity data is input into the trained trend prediction model, and the influence of viscosity on the corrosion of heat exchanger 3 is inferred based on the viscosity change trend output by the model; the model will judge the potential influence on the corrosion degree of heat exchanger 3 based on the change amplitude and trend of viscosity data. High viscosity may usually lead to increased deposits, while too low viscosity may affect fluidity and make corrosive media more easily attached; the model can identify whether the change of real-time viscosity data may accelerate corrosion, providing a basis for timely decision-making and adjustment of operating parameters.

[0150] Based on the corrosion identification results, the viscosity change trend in the future period is analyzed, and the corrosion trend of the heat exchanger 3 is predicted, including whether the corrosion rate will increase and whether the corrosion area will expand. Predicting the corrosion trend in advance is helpful for the formulation of maintenance plans, and the operating parameters can be adjusted in advance according to potential corrosion risks, thereby extending the service life of the heat exchanger 3 and avoiding production interruptions and equipment failures.

[0151] By analyzing the viscosity trend, the effect of viscosity on corrosion efficiency (i.e. the effect of viscosity change on corrosion rate) can be quantified. Combined with the trend prediction model, it can be estimated when and under what conditions corrosion may intensify or reach a time point where maintenance is required; quantified corrosion efficiency and time points provide managers with clear reference indicators of corrosion degree, making it easier to formulate detailed equipment inspection and maintenance strategies to avoid accidental equipment damage.

[0152] According to the predicted corrosion time node, the maintenance and care work of the heat exchanger 3 is arranged in advance to avoid the impact of excessive corrosion or reduced heat exchange efficiency on the production process; through preventive maintenance, the downtime of the heat exchanger 3 is reduced, the production continuity is improved, and the maintenance cost is effectively reduced. At the same time, the service life of the heat exchanger 3 is extended and the utilization efficiency of resources is optimized.

[0153] In a preferred embodiment of the present invention, the above trend prediction model includes:

[0154] The historical viscosity data is encoded into sequence data, and the sequence data is input into the trend prediction model; the trend prediction model includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the intermediate representation data of the hidden layer is transmitted to the output layer, and the output layer output represents the corrosion degree recognition result of the viscosity data, which is as follows:

[0155] , where W u , W r , W c represents the weight parameter, b u 、b r 、b c represents the bias parameter, represents the dot product, u （t） 、r （t） and c （t） Represent the intermediate states of the first, second, and third hidden layers respectively, X （t） The data item representing the viscosity data at the tth moment, H （t） and H （t-1） They represent the intermediate representation data of the viscosity data at the t-th and t-1-th moments, n ≥ t ≥ 1, n represents the total number of training input viscosity data, and H （t-1） =X （t） , tanh is the hyperbolic tangent function, σ represents the sigmoid function;

[0156] Output layer: ; Among them, y t represents the output vector of the t-th intermediate period, where the components of the output vector represent the input data item belonging to the i-th state, and h t represents the tth intermediate representation data, W y is the weight parameter, b y is the bias parameter, σ represents the sigmoid function. The output represents the corrosion degree recognition result of viscosity data;

[0157] Based on the constructed trend prediction model, the acquired real-time viscosity data is input into the trend prediction model to output the corrosion degree identification result of the real-time viscosity data;

[0158] The corrosion degree identification result of the real-time viscosity data is compared with the corrosion degree identification result of the historical viscosity data to determine the development trend of the corrosion degree of the heat exchanger 3 due to the viscosity change.

[0159] In this embodiment, the historical viscosity data is converted into a time series format to meet the input requirements of the prediction model. The time series data contains the values ​​of viscosity at different time points, which can reflect the dynamic changes of viscosity over time; the historical data provided in the form of time series enables the model to identify the long-term trend of viscosity changes, thereby more accurately predicting future data.

[0160] The model consists of an input layer, a first hidden layer, a second hidden layer, a third hidden layer, and an output layer. The input layer receives time series data, the hidden layer is used to extract data features, and the output layer finally gives the corrosion identification results.

[0161] Functions of each layer:

[0162] Input layer: Receives the encoded viscosity data sequence and passes it to the hidden layer step by step.

[0163] Hidden layer: nonlinear transformation of data through weight and bias parameters. The intermediate state of each layer (u (t) ,r (t) ,c (t) ) represents the feature representation of different layers, which is used to extract the viscosity change characteristics more deeply.

[0164] Output layer: Integrate the intermediate representation data of the hidden layer and output the corrosion degree recognition result of the viscosity data.

[0165] The hierarchical structure ensures that the model can perform multi-level analysis and extraction of different features of viscosity data, thereby obtaining a more accurate prediction of the corrosion degree.

[0166] Hidden layer parameters: weight parameter (W u ,W r ,W c ) and the bias parameter (b u ,b r ,b c ) controls the output of each layer, and the dot product operation combines the weights and input data.

[0167] Status: u (t) ,r (t) and c (t) They represent the intermediate states of the hidden layers respectively, and extract features by transforming the input data; the model transforms the input data X at each time step t (t) Process and output real-time status H (t) At the initial time t=1, H (t-1) =X (t) It means that the input data is directly used as the initial state and is gradually updated with time steps.

[0168] Activation function:

[0169] Tanh function: The hyperbolic tangent function is used for nonlinear mapping in the hidden layer to help capture the complex characteristics of the input data.

[0170] σ function (sigmoid): The activation function is used to normalize the input data, compress the data within the range of (0,1), and increase the expressiveness of the model.

[0171] Through the use of multi-layer hidden layer structure and nonlinear activation function, the model can better learn the temporal patterns and characteristics of viscosity data to ensure the accuracy of corrosion degree identification.

[0172] After obtaining the real-time viscosity data, it is input into the trained trend prediction model, and the model will predict the degree of corrosion based on the real-time data; it can judge in real time the impact of the real-time fluid viscosity on the degree of corrosion of the heat exchanger 3, and provide early warning to prevent the corrosion from worsening.

[0173] The corrosion identification results of the real-time viscosity data are compared with the identification results of the historical data to determine whether the viscosity change accelerates the development of corrosion or slows down the corrosion process; if the real-time corrosion level is significantly different from the historical results, it indicates that the viscosity change may cause corrosion to increase or slow down, and the operating parameters can be further analyzed and optimized; by comparing the historical and real-time data, the corrosion condition of the heat exchanger 3 can be dynamically monitored, and it can be judged in time whether measures need to be taken, so as to effectively control the corrosion risk.

[0174] In a preferred embodiment of the present invention, the estimated corrosion efficiency and corrosion time node include:

[0175] Based on the development trend, according to the real-time fluid viscosity, fluid composition, fluid velocity, pulse width and power shock wave data, through: ; to obtain the corrosion rate; R represents the corrosion rate; k represents the corrosion rate constant; E q represents the fluid flow rate; T represents the temperature; F represents the ideal fluid component category parameter set; e represents the power shock wave data; R T Indicates corrosion temperature;

[0176] When the allowable corrosion depth of the heat exchanger 3 is d, the corrosion time node is estimated according to the corrosion rate, by: ; To obtain the corrosion time node t;

[0177] When the estimated corrosion time node t is lower than the preset safety time, the trend prediction module issues an alarm.

[0178] In this embodiment, the corrosion rate is obtained based on the real-time fluid viscosity, composition, flow rate, pulse width and power shock wave data; by integrating multiple factors affecting corrosion, the model can comprehensively evaluate the corrosion rate under real-time conditions. In particular, factors such as fluid viscosity, fluid composition, and flow rate directly affect the contact frequency and degree of corrosion between the fluid and the equipment surface, while pulse width and power shock wave data may affect the impact strength of the fluid.

[0179] The corrosion rate constant k is a predefined parameter that usually depends on the corrosion resistance and working environment. R represents the ideal fluid component category parameter set, that is, an ideal corrosion benchmark that can be used to correct the real-time corrosion rate; the addition of the corrosion rate constant k and the ideal component category parameter set R makes the model more applicable and adjustable. These two parameters take into account the material properties and ideal corrosion state, and can effectively compare the deviation between the real-time and ideal conditions, and assist in accurate prediction.

[0180] The expected corrosion time node t can be calculated by the classification parameter set R and the allowable corrosion depth d of the heat exchanger 3 material: ; In the formula, t represents the corrosion time node, d represents the allowable corrosion depth, and R is the ideal fluid component category parameter set; the purpose of corrosion time node calculation is to evaluate how long the equipment material can withstand the real-time corrosion rate and ensure that the heat exchanger 3 material operates safely without exceeding the allowable corrosion depth. This estimated time node provides a clear time indicator, which helps to plan maintenance and implement preventive measures.

[0181] When the predicted corrosion time node t is lower than the preset safety time, the trend prediction module will automatically issue an alarm. The preset safety time represents the shortest safety time under normal operation of the heat exchanger 3; the alarm system provides an advance notification mechanism for operators to ensure that necessary measures can be taken before the corrosion threat becomes serious to avoid emergency shutdowns caused by corrosion problems. The early warning mechanism can help optimize maintenance arrangements and minimize equipment downtime and operational risks.

[0182] like Figure 2 As shown, a heat exchanger descaling and anti-corrosion device based on intelligent control is used to implement a heat exchanger descaling and anti-corrosion system based on intelligent control, and the device includes:

[0183] Intelligent digital electric control cabinet 1, energy conversion probe 2, heat exchanger 3, link wire 4, digital instrument wire 5 and digital intelligent display screen 6;

[0184] One end of the link wire 4 is connected to the intelligent digital electric control cabinet 1, and the other end is connected to the energy conversion probe 2; the intelligent digital electric control cabinet 1 sends a telecommunication signal which is transmitted to the energy conversion probe 2 through the link wire 4, so that the telecommunication signal is converted into a power oscillation wave; the power oscillation wave automatically removes scale and prevents corrosion on the heat exchanger 3;

[0185] The energy conversion probe 2 is arranged on the heat exchanger 3 and is tightly connected to the heat exchanger 3; a plurality of plates and a plurality of tubes are arranged in the heat exchanger 3, and the tubes are arranged through the plates. The energy conversion probe 2 transmits the power shock wave to each plate and each tube. The power shock wave runs at a high speed in the plates and tubes so that the particles in the plates collide with each other to eliminate stress corrosion; and micro-impact is generated on the inside and outside of the plates and tubes to prevent scaling from approaching and adhering to the heat exchanger 3;

[0186] The intelligent digital electric control cabinet 1 is provided with a viscosity meter, which is connected to the heat exchanger 3 via a digital instrument wire 5 to detect the fluid viscosity in the heat exchanger 3; the digital intelligent display screen 6 is provided on the intelligent digital electric control cabinet 1 to display the fluid viscosity data;

[0187] The fluid viscosity data is transmitted to the remote mobile phone control terminal through the chip, and the remote mobile phone control terminal remotely adjusts the pulse width and period of multiple gears of the power oscillation wave corresponding to the fluid viscosity to achieve the best automatic descaling and anti-corrosion effect.

[0188] The device is composed of an intelligent digital electric control cabinet 1, an energy conversion probe 2, a heat exchanger 3, a link wire 4, a digital instrument wire 5 and a digital intelligent display screen 6. The energy conversion probe 2 is tightly connected to the heat exchanger 3 at an effective position by different methods such as threading or welding according to the situation. The power shock wave is quickly transmitted to each plate and each tube of the heat exchanger 3, runs at a high speed in the tube plate, and generates micro-impact on the inner and outer sides of the tube plate, which effectively prevents the approach and adhesion of scale, cleans the wall surface, and eliminates the occurrence of under-scale corrosion. The power shock wave causes the particles in the plate to collide with each other, thereby eliminating stress corrosion.

[0189] One of the present invention is that the iron core of the energy conversion probe 2 adopts a new type of magnetic material independently developed and containing 60% cobalt, 3% vanadium, and the balance iron. Without heat treatment, the energy conversion rate of existing domestic products can be increased by more than 30%, the temperature of the energy conversion probe 2 can be reduced by more than 25%, the intensity of the power shock wave can be increased by more than 30%, and the automatic descaling and anti-corrosion effect of the heat exchanger 3 can be greatly enhanced.

[0190] The second feature of the present invention is that it adopts a fully intelligent digital remote control system. A viscosity meter is installed in the intelligent digital electric control cabinet 1, which can perform online digital detection of the viscosity of the fluid in the heat exchanger 3. The viscosity data is transmitted to a remote mobile phone through a chip. According to the provided viscosity data (the viscosity is set to five gears: low viscosity, medium viscosity, medium viscosity, high viscosity, and high viscosity), the pulse width and period of the five gears corresponding to the power oscillation wave are remotely controlled, so that the device can achieve the best automatic descaling and anti-corrosion effect on the heat exchanger 3.

[0191] The third feature of the present invention is that the high-intensity power wave generated by the device can eliminate the flow layer in the heat exchanger 3, turning heat conduction into convection heat transfer, and can improve the heat exchange efficiency by more than 10%.

[0192] According to statistics, the energy loss caused by scaling and corrosion of heat exchangers 3 in my country accounts for more than 0.4% of the national income each year. Chemical cleaning not only damages the equipment, but also pollutes the environment. Leakage and explosion caused by corrosion also occur from time to time, causing significant losses to life and property. This device can make the heat exchanger 3 scale-free, and can play a huge role in energy saving and carbon reduction, emission reduction and efficiency improvement, environmental protection and safety, etc., and can be called the fifth energy!

[0193] When the functions of the above modules are implemented in the form of software functional units and used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks or optical disks.

[0194] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general intelligent device. Therefore, the purpose of the present invention can also be achieved by simply providing a program product containing a program code that implements the method or system. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but it is not necessary to perform them in chronological order. Some steps can be performed in parallel or independently of each other.

Claims

1. A heat exchanger descaling and anti-corrosion system based on intelligent control, characterized in that: include: A real-time monitoring module: used for real-time monitoring of key parameters of the heat exchanger (3); the key parameters include the temperature, flow rate and water quality component parameters of the heat exchanger (3); An abnormality recognition module: it is used to collect abnormal image data based on the abnormal key parameters of the heat exchanger (3) monitored by the real-time monitoring module; construct a training set with the abnormal image data of the heat exchanger (3) to obtain an abnormality recognition model, so as to output abnormality type data; Inputting the abnormal image data collected in real time into an abnormality recognition model to output a recognition result of the abnormal type data of the real-time heat exchanger (3); and judging the scaling thickness based on the recognition result; A control module: it is used to obtain real-time scaling thickness information data based on the real-time abnormal type data identification result of the heat exchanger (3), and automatically perform scaling and corrosion prevention treatment on the heat exchanger (3) by self-adjusting the pulse width and period of the power oscillation wave according to the real-time acquired fluid viscosity data, so as to prevent scaling adhesion; and eliminate stress corrosion of the heat exchanger (3); A corrosion prediction module: used to adjust the power oscillation wave pulse width and period results based on the control module to predict the corrosion rate and corrosion time of the heat exchanger (3) to prevent the heat exchanger (3) from being affected by corrosion and affecting the operation of the equipment; Wherein, the real-time monitoring module further includes: Monitoring unit, collection unit and pre-processing unit; The monitoring unit monitors the temperature, flow rate and water quality component parameter data of the heat exchanger (3) in real time, sets a standard threshold value for each parameter, and when the monitored parameter data is greater than the standard threshold value, it indicates that the parameter data is abnormal; Based on the monitored parameter abnormal data, the acquisition unit acquires the parameter abnormal image data of the heat exchanger (3) to obtain the abnormal image data; Based on the abnormal image data, the preprocessing unit preprocesses the abnormal image data to obtain clear abnormal image data; Wherein, the abnormality identification module further includes: Analysis unit, recognition unit and judgment unit; The analyzing unit extracts features from the acquired clear abnormal image data to obtain feature image data; Inputting the characteristic image data of the heat exchanger (3) history into the recognition unit to train and construct the abnormality recognition model; the abnormality recognition model outputs the recognition result of the abnormality type data of the heat exchanger (3); Based on the identification result of the abnormal type data, the judgment unit judges the scaling thickness information data through the scaling type characteristic data; Wherein, based on the identification result of the abnormal type data, the judgment unit obtains scaling type characteristic data to judge scaling thickness information data, including: According to the recognition result, the judgment unit obtains the characteristic data of the scaling area, wherein the characteristic data includes the shape, color and distribution area data of the scaling; Based on the determination unit, the scaling area characteristic data is obtained, and the scaling thickness information data is obtained by extracting the RGB color mode data of the scaling area, analyzing the color chromaticity C, brightness L and hue θ parameters of the scaling area, and combining the scaling density information data D. ; Where f is the conversion function.

2. The heat exchanger descaling and anti-corrosion system based on intelligent control according to claim 1, characterized in that: The control module further includes: Based on the real-time scaling thickness information data, obtaining heat exchange efficiency data of the heat exchanger (3); Based on the scaling thickness information data and the heat exchange efficiency data of the heat exchanger (3), the real-time fluid viscosity data Deviation analysis is performed to obtain the impact data of the deviation between fluid viscosity and heat exchange demand of the heat exchanger (3); where μ represents the real-time fluid viscosity; SR (t) is the time viscosity reading of the viscosity sensor at the tth moment; Based on the impact data, a preset threshold is set, and the impact data is compared with the preset threshold; based on the comparison result, the control module self-adjusts the pulse width and period of the power oscillation wave to obtain an automatic descaling and anti-corrosion strategy for the heat exchanger (3).

3. The heat exchanger descaling and anti-corrosion system based on intelligent control according to claim 2, characterized in that: Based on the influence data, a preset threshold is set, and the influence data is compared with the preset threshold; Based on the comparison result, the control module self-adjusts the pulse width and period of the power oscillation wave to obtain an automatic descaling and anti-corrosion strategy for the heat exchanger (3), and further includes: According to the comparison result of the influence degree data, when the influence degree data is greater than or equal to the preset threshold, the real-time fluid viscosity data μ is too high. By increasing the power shock wave intensity by more than 30%, the wavelength is increased to impact the scaling layer with tight adhesion. Based on increasing the power shock wave intensity, through: ; Increase the pulse width to 300μs so that the power shock wave can further impact the scale until it falls off; and automatically increase the disturbance effect of the power shock wave to reduce the contact time between the fluid corrosive medium and the metal surface; where w represents the real-time shock wave pulse width; w0 represents the initial set pulse width; K w is the pulse width adjustment coefficient, which indicates the sensitivity of pulse width to viscosity change; μ opt Indicates optimum or target viscosity; When the impact data is less than the preset threshold, the real-time fluid viscosity data μ is too low, through: ; Increase the power oscillation frequency by 20-40Hz; and reduce the pulse width by 100μs to allow the fluid to circulate more quickly and reduce the retention of corrosive media; where g represents the real-time oscillation frequency; g0 represents the initial set oscillation frequency; K g Indicates the frequency adjustment coefficient, which indicates the sensitivity of frequency to viscosity changes; μ opt Indicates optimum or target viscosity; According to the viscosity change through: ; To adjust the control parameters of the power oscillation wave so that the control module tends to a stable state; where u(t) is the control signal at the tth moment, representing the control signal for adjusting the oscillation wave frequency or pulse width; K p Represents proportional gain; K i represents differential gain; K d represents the interaction gain; represents the cumulative deviation; Indicates the rate of change of the deviation.

4. The heat exchanger descaling and anti-corrosion system based on intelligent control according to claim 3, characterized in that: The corrosion prediction module further includes: Acquiring historical viscosity data and training to construct a trend prediction model; collecting real-time viscosity data of the fluid; inputting the real-time viscosity data into the trend prediction model to output a result of identifying the corrosion degree of the heat exchanger (3) based on the viscosity data; and predicting the corrosion trend of the heat exchanger (3) based on the identification result; Based on the trend prediction model, the changing trend of the corrosion effect of the viscosity data on the heat exchanger (3) is predicted to quantify the corrosion efficiency and corrosion time node of the fluid viscosity on the heat exchanger (3); and based on the corrosion time node, the heat exchanger (3) is maintained in advance.

5. The heat exchanger descaling and anti-corrosion system based on intelligent control according to claim 4, characterized in that: Acquiring historical viscosity data and training to construct a trend prediction model; collecting real-time viscosity data of the fluid; inputting the real-time viscosity data into the trend prediction model to output a result of identifying the degree of corrosion of the heat exchanger (3) based on the viscosity data; Based on the identification results, the corrosion trend of the heat exchanger (3) is predicted, including: The historical viscosity data is encoded into sequence data, and the sequence data is input into the trend prediction model; the trend prediction model includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the intermediate representation data of the hidden layer is transmitted to the output layer, and the output layer output represents the corrosion degree recognition result of the viscosity data, which is as follows: , where W u , W r , W c represents the weight parameter, b u , b r , b c represents the bias parameter, represents the dot product, u （t） 、r （t） and c （t） Represent the intermediate states of the first, second, and third hidden layers respectively, X （t） The data item representing the viscosity data at the tth moment, H （t） and H （t-1） They represent the intermediate representation data of the viscosity data at the t-th and t-1-th moments, n ≥ t ≥ 1, n represents the total number of training input viscosity data, and H （t-1） =X （t） , tanh is the hyperbolic tangent function, σ represents the sigmoid function; Based on the constructed trend prediction model, the acquired real-time viscosity data is input into the trend prediction model to output the corrosion degree identification result of the real-time viscosity data; The corrosion degree identification result of the real-time viscosity data is compared with the corrosion degree identification result of the historical viscosity data to determine the development trend of the viscosity change on the corrosion degree of the heat exchanger (3).

6. The heat exchanger descaling and anti-corrosion system based on intelligent control according to claim 5, characterized in that: Predicting the changing trend of the corrosion effect of the viscosity data on the heat exchanger (3) based on the trend prediction model, so as to quantify the corrosion efficiency and corrosion time node of the fluid viscosity on the heat exchanger (3); According to the corrosion time node, the heat exchanger (3) is maintained in advance, including: Based on the development trend, according to the real-time fluid viscosity, fluid composition, fluid velocity, pulse width and power shock wave data, through: ; to obtain the corrosion rate; R represents the corrosion rate; k represents the corrosion rate constant; E q represents the fluid flow rate; T represents the temperature; F represents the ideal fluid component category parameter set; e represents the power shock wave data; R T Indicates corrosion temperature; When the allowable corrosion depth of the heat exchanger (3) is d, the corrosion time node is estimated according to the corrosion rate, by: ; To obtain the corrosion time node t; When the estimated corrosion time node t is lower than the preset safety time, the corrosion prediction module issues an alarm.

7. A heat exchanger descaling and anti-corrosion device based on intelligent control, characterized in that: The device is used to implement the heat exchanger descaling and anti-corrosion system based on intelligent control as described in any one of claims 1 to 6, and the device includes: Intelligent digital electric control cabinet (1), energy conversion probe (2), heat exchanger (3), link wire (4), digital instrument wire (5) and digital intelligent display screen (6); One end of the link wire (4) is connected to the intelligent digital electric control cabinet (1), and the other end is connected to the energy conversion probe (2); the intelligent digital electric control cabinet (1) sends a telecommunication signal which is transmitted to the energy conversion probe (2) via the link wire (4), so that the telecommunication signal is converted into a power oscillation wave; the power oscillation wave automatically removes scale and prevents corrosion on the heat exchanger (3); The energy conversion probe (2) is arranged on the heat exchanger (3) and is tightly connected to the heat exchanger (3); a plurality of plates and a plurality of tubes are arranged in the heat exchanger (3), and the tubes are arranged through the plates. The energy conversion probe (2) transmits the power shock wave to each plate and each tube. The power shock wave runs at a high speed in the plates and tubes so that particles in the plates collide with each other to eliminate stress corrosion; and micro-impact is generated on the inner and outer sides of the plates and tubes to prevent scaling from approaching and adhering to the heat exchanger (3); The intelligent digital electric control cabinet (1) is provided with a viscosity meter, and the viscosity meter is connected to the heat exchanger (3) via a digital instrument wire (5) to detect the viscosity of the fluid in the heat exchanger (3); the digital intelligent display screen (6) is provided on the intelligent digital electric control cabinet (1) to display the fluid viscosity data; The fluid viscosity data is transmitted to the remote mobile phone control terminal through the chip, and the remote mobile phone control terminal remotely adjusts the pulse width and period of multiple gears of the power oscillation wave corresponding to the fluid viscosity to achieve the best descaling and anti-corrosion effect.

Images