Boiler wall temperature prediction system and method based on digital twinborn model

Abstract

The invention discloses a boiler wall temperature prediction system and method based on a digital twinborn model. The system comprises a data acquisition module which is connected with a data processing and analyzing module and is used for acquiring operation data of a boiler in real time and transmitting the operation data to the data processing and analyzing module; the data and analysis module is connected with the digital twinborn model module and used for preprocessing the operation data to obtain target data and transmitting the target data to the digital twinborn model module; the digital twinborn model module is used for constructing a boiler digital twinborn model according to the operation data and the target data; and the prediction and early warning module is connected with the digital twinborn model module and is used for predicting the wall temperature of the boiler digital twinborn model in the digital twinborn model module by utilizing a machine learning algorithm to obtain a prediction result and carrying out fault diagnosis and early warning according to the prediction result. According to the technical scheme, high-precision and real-time simulation and prediction of the boiler wall temperature are achieved, and the safety and economical efficiency of boiler operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of boiler wall temperature prediction, and particularly to a boiler wall temperature prediction system and method based on a digital twin model. Background Art

[0002] A boiler is an energy conversion device. The energy input into the boiler includes the chemical energy in fuel and electric energy, and the boiler outputs steam, high-temperature water or organic heat carrier with a certain amount of heat energy.

[0003] In the field of boiler wall temperature simulation and prediction, certain progress has been made in the prior art, but there are still some defects and deficiencies, mainly including the limitations of numerical simulation methods, the deficiencies of data-driven models, and the lack of integration of the prior art. Among them, the limitations of numerical simulation methods include high computational complexity: The calculation process of traditional numerical simulation methods is complex and the amount of calculation is large, making it difficult to meet the requirements of real-time monitoring and prediction. Model simplification leads to insufficient accuracy: Due to the huge size difference between the boiler furnace and the heating surface pipes, most numerical models can only focus on either the flue gas side or the working medium side of the boiler, and a large number of assumptions or simplifications affect the key parameters on the other side that affect the wall temperature, seriously affecting the accurate prediction of the wall temperature distribution. Inability to cover all working conditions: Numerical simulation methods usually can only calculate for limited working conditions and are difficult to cover all working conditions during boiler operation, especially the heat transfer deterioration problem under low load or variable load conditions. The deficiencies of data-driven models include data quality and quantity limitations: The application of data-driven models in wall temperature distribution prediction is difficult, mainly because it is difficult to measure the wall temperature of the pipe and the data is usually discontinuous, resulting in the quantity and quality of the data required for model training not being able to meet the requirements.

[0004] Poor adaptability to complex working conditions: The existing data-driven models have a low prediction accuracy for the heat transfer deterioration problem under low load conditions because these models lack sufficient heat transfer deterioration characteristic data, resulting in insufficient extraction ability for this characteristic. Limited model generalization ability: Although data-driven models can achieve the generalization of numerical simulation results for limited working conditions, when facing complex working condition changes, the generalization ability of the models is still insufficient and it is difficult to adapt to various emergencies during boiler operation. The lack of integration of the prior art includes lack of real-time performance: In the prior art, although there are some methods that can achieve the prediction of boiler wall temperature, most of them cannot meet the real-time requirement and cannot provide timely decision support for boiler operators. Lack of fault diagnosis and early warning functions: The prior art mainly focuses on the prediction of wall temperature and lacks the functions of diagnosing and warning abnormal situations of boiler wall temperature, and cannot detect potential safety hazards in advance. Summary of the Invention

[0005] The present invention provides a boiler wall temperature prediction system and method based on a digital twin model, which solves the defects existing in the existing boiler wall temperature monitoring and prediction technologies, overcomes the deficiencies of the existing technologies, realizes high-precision and real-time simulation and prediction of the boiler wall temperature, and improves the safety and economy of boiler operation.

[0006] According to one aspect of the present invention, there is provided a boiler wall temperature prediction system based on a digital twin model, the system comprising: a data acquisition module, a data processing and analysis module, a digital twin model module, and a prediction and warning module; wherein,

[0007] The data acquisition module, connected to the data processing and analysis module, is configured to collect the operation data of the boiler in real time and transmit the operation data to the data processing and analysis module; wherein, the operation data includes sensor data, historical operation data, and real-time operation parameters of the boiler;

[0008] The data and analysis module, connected to the digital twin model module, is configured to preprocess the operation data to obtain target data and transmit the target data to the digital twin model module;

[0009] The digital twin model module is configured to construct a boiler digital twin model according to the operation data and the target data;

[0010] The prediction and warning module, connected to the digital twin model module, is configured to predict the wall temperature of the boiler digital twin model in the digital twin model module by using a machine learning algorithm to obtain a prediction result, and perform fault diagnosis and warning according to the prediction result.

[0011] According to another aspect of the present invention, there is provided a boiler wall temperature prediction method based on a digital twin model, the method comprising:

[0012] Collect the operation data of the boiler in real time; wherein, the operation data includes sensor data, historical operation data, and real-time operation parameters of the boiler;

[0013] Preprocess the operation data to obtain target data;

[0014] Construct a boiler digital twin model according to the operation data and the target data;

[0015] Predict the wall temperature of the boiler digital twin model by using a machine learning algorithm to obtain a prediction result, and perform fault diagnosis and warning according to the prediction result.

[0016] The technical solution of the embodiment of the present invention realizes the real-time simulation and accurate prediction of the boiler wall temperature through a digital twin model, solves the defects existing in the existing boiler wall temperature monitoring and prediction technology, overcomes the deficiencies of the existing technology, and improves the safety and economy of boiler operation.

[0017] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 is a schematic diagram of a boiler wall temperature prediction system based on a digital twin model provided in Embodiment 1 of the present invention;

[0020] Figure 2 is a flowchart of a boiler wall temperature prediction method based on a digital twin model provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0022] It should be noted that the terms "target" and the like in the specification and claims of the present invention are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0023] Embodiment 1

[0024] Figure 1 It is a schematic diagram of a boiler wall temperature prediction system based on a digital twin model provided by Embodiment 1 of the present invention. As Figure 1 shown, the system includes: a data acquisition module 110, a data processing and analysis module 120, a digital twin model module 130, and a prediction and warning module 140; wherein,

[0025] The data acquisition module 110 is connected to the data processing and analysis module 120 and is used to collect the operation data of the boiler in real time and transmit the operation data to the data processing and analysis module 120; wherein, the operation data includes sensor data, historical operation data, and real-time operation parameters of the boiler;

[0026] The data and analysis module 120 is connected to the digital twin model module 130 and is used to preprocess the operation data to obtain target data and transmit the target data to the digital twin model module 130;

[0027] The digital twin model module 130 is used to construct a boiler digital twin model based on the operation data and target data;

[0028] The prediction and warning module 140 is connected to the digital twin model module 130 and is used to predict the wall temperature of the boiler digital twin model in the digital twin model module by using a machine learning algorithm to obtain a prediction result, and perform fault diagnosis and warning according to the prediction result.

[0029] In this solution, the sensor data includes temperature sensor data, pressure sensor data, flow sensor data, etc.

[0030] Among them, the historical operation data refers to the operation data of the boiler at historical moments. The historical operation data includes historical sensor data and historical real-time operation parameters of the boiler.

[0031] In this embodiment, during the operation of the boiler, the real-time operation parameters of the boiler mainly refer to the parameters that need to be monitored and adjusted in real time during the operation of the boiler. The real-time operation parameters of the boiler include combustion-related parameters, steam-related parameters, water-related parameters, and other operation parameters, etc. Among them, the combustion-related parameters include the total coal amount, the air volume of the burner, and the burnout air ratio. The total coal amount refers to the total amount of coal burned by the boiler per unit time, usually expressed in tons per hour (t / h). Adjusting the total coal amount in real time can control the combustion intensity and ensure that the output power of the boiler matches the load demand. The air volume of the burner refers to the air volume of each burner, with the unit of kilograms per second (kg / s). By adjusting the air volume of the burner, the combustion efficiency can be optimized and pollutant emissions can be reduced. The burnout air ratio refers to the air volume used to control the later stage of combustion to ensure that the fuel is fully burned and reduce the loss of incomplete combustion. The steam-related parameters include steam pressure, steam temperature, and steam mass flow rate. The steam pressure refers to the steam pressure generated by the boiler, with the unit of megapascals (MPa). Monitoring the steam pressure in real time can ensure that the boiler operates within a safe pressure range. The steam temperature refers to the temperature of the steam, with the unit of degrees Celsius (°C) or Kelvin (K). By controlling the steam temperature, the efficiency of the steam turbine can be optimized. The steam mass flow rate refers to the mass flow rate of the steam passing through the boiler per unit time, with the unit of tons per hour (t / h). Adjusting the steam flow rate in real time can ensure that the output of the boiler matches the system demand. The water-related parameters include feed water flow rate, the water temperature at the outlet of the economizer, and the drum water level. The feed water flow rate refers to the amount of water entering the boiler per unit time, with the unit of tons per hour (t / h). Monitoring and adjusting the feed water flow rate in real time can ensure the stability of the boiler water level. The water temperature at the outlet of the economizer refers to the water temperature at the outlet of the economizer, with the unit of degrees Celsius (°C). By controlling the water temperature at the outlet of the economizer, the thermal efficiency of the boiler can be optimized. The drum water level refers to the height of the water level in the boiler drum, with the unit of millimeters (mm). Monitoring the drum water level in real time can prevent the boiler from drying out or overfilling. Other operation parameters include the flue gas temperature, the burner configuration mode, and the furnace negative pressure. The flue gas temperature refers to the temperature of the flue gas discharged from the boiler, with the unit of degrees Celsius (°C). By monitoring the flue gas temperature, the combustion efficiency and heat loss can be judged. The burner configuration mode refers to the operation mode of the burner, for example, which burners are in operation and which are in standby state. By adjusting the burner configuration, the combustion process can be optimized. The furnace negative pressure refers to the pressure inside the furnace, usually expressed in pascals (Pa). By controlling the furnace negative pressure, the stability and safety of the combustion process can be ensured.

[0032] Specifically, the data acquisition module 110 collects various types of data during the operation of the boiler in real time, including sensor data, historical operation data, and the real-time operation parameters of the boiler.

[0033] Optionally, the data acquisition module 110 includes a sensor network unit, a data transmission unit, and a data storage unit; among them,

[0034] The sensor network unit is used to collect the operation data of the boiler in real time based on pre-arranged sensors;

[0035] The data transmission unit is used to transmit the operation data to the data processing and analysis module 120 through industrial Ethernet or wireless communication technology;

[0036] The data storage unit is used to store the operation data in a distributed database.

[0037] In this solution, the sensors in the sensor network unit include temperature sensors, pressure sensors, and flow sensors. Each sensor is arranged on the boiler wall. The operation data of the boiler is collected in real time by each sensor.

[0038] In this embodiment, based on the data transmission unit, the operation data in the SIS system is transmitted to the data processing and analysis module 120 through industrial Ethernet or wireless communication technology.

[0039] Furthermore, the data storage unit stores the operation data in a distributed database and supports the efficient storage and fast query of large-scale data.

[0040] Through multi-source data fusion technology, combining sensor data, historical operation data, and real-time operation parameters of the boiler, it comprehensively reflects the operation state of the boiler, can achieve high-precision simulation and prediction of the boiler wall temperature, and improves the real-time performance and accuracy of wall temperature prediction.

[0041] In this solution, preprocessing mainly refers to the process of collecting, cleaning, transforming, etc. the original data to improve the data quality, enhance the performance of the algorithm model, and make it more suitable for subsequent analysis and modeling tasks.

[0042] Specifically, the data and analysis module 120 preprocesses the operation data to obtain target data and provides input for the digital twin model module 130.

[0043] Optionally, the data processing and analysis module 120 includes a data preprocessing unit and a feature extraction unit; among them,

[0044] The data preprocessing unit is used to clean, denoise, and normalize the operation data to obtain preprocessed data;

[0045] The feature extraction unit is used to extract features from the preprocessed data to obtain target data.

[0046] In this embodiment, data cleaning mainly refers to discovering and correcting recognizable errors in data files, including checking data consistency, handling invalid values and missing values, etc. Data denoising refers to removing noise from the collected data, restoring the true appearance of the data, and improving data quality for subsequent analysis and processing. Data normalization is to map data to a specific interval or range according to a certain method, so that the data has a unified scale and distribution, and eliminates the influence of dimensions and orders of magnitude between different features. By cleaning, denoising, and normalizing the operation data, the data quality can be improved.

[0047] In this solution, machine learning algorithms can be used to extract the wall temperature-related features from the preprocessed data, and the feature variables with high correlation degrees are screened out through the grey relational analysis method, and the feature variables are used as the target data. Among them, the target data includes combustion efficiency, flue gas temperature, working medium flow rate, etc.

[0048] By performing preprocessing and feature extraction operations on the operation data, the quality of the data can be improved.

[0049] Optionally, the data processing and analysis module 120 further includes a model training unit; among them,

[0050] The model training unit is used to train the deep learning model to be trained according to the target data to obtain a deep learning model.

[0051] In this solution, based on historical target data and real-time target data, the deep learning model is trained to improve the prediction accuracy of the model. Among them, the deep learning model includes DNN (Deep Neural Networks, deep neural network) and LSTM (Long Short-Term Memory, long short-term memory network).

[0052] In this embodiment, the digital twin model module 130 is used to construct a boiler digital twin model according to the operation data and target data, and perform model dynamic update and simulation in combination with the data. Specifically, the digital twin technology can be used to construct a boiler digital twin model according to the operation data and target data. Digital twin is a simulation process that makes full use of data such as physical models, sensors, and operation history, integrates multiple disciplines, multiple physical quantities, multiple scales, and multiple probabilities, and completes the mapping in the virtual space, so as to reflect the full life cycle process of the corresponding physical equipment.

[0053] Optionally, the digital twin model module 130 includes a model construction unit and a real-time update unit; among them,

[0054] The model construction unit is used to construct a boiler digital twin model according to the operation data and target data;

[0055] The real-time update unit is used to update the operation parameters of the boiler digital twin model in real time.

[0056] Specifically, based on the physical model, operation data, and target data of the boiler, a boiler digital twin model is constructed. Among them, the boiler digital twin model includes the geometric structure, combustion process, heat transfer process, etc. of the boiler. Constructing a boiler digital twin model that includes multiple physical fields such as the geometric structure, combustion process, and heat transfer process of the boiler can comprehensively reflect the physical process of the boiler operation.

[0057] Furthermore, through the data provided by the data processing and analysis module 120, the boiler digital twin model is dynamically updated to ensure that the model can reflect the actual operation state of the boiler in real time. Combining sensor data, historical operation data, and real-time operation parameters of the boiler improves the accuracy and reliability of the model.

[0058] The operation parameters of the boiler digital twin model are updated in real time to ensure that the model always reflects the actual operation state of the boiler, thereby improving the accuracy and timeliness of prediction.

[0059] Furthermore, the boiler digital twin model is not only a tool for data collection and storage, but also the core for realizing data processing, analysis, prediction, and early warning. Through real-time update and dynamic adjustment, combined with multi-source data fusion and intelligent analysis, it provides comprehensive technical support for the safe and efficient operation of the boiler.

[0060] Specifically, the boiler digital twin model obtains various types of data during the operation of the boiler in real time through connection with the sensor network. These data include sensor data such as temperature, pressure, and flow rate, as well as real-time operation parameters and historical operation data of the boiler. Through the virtual mapping of the digital twin model, these data can be managed and integrated more efficiently.

[0061] Furthermore, through multi-physical field coupling simulations (such as combustion process, heat transfer process) and machine learning algorithms, the collected data is deeply analyzed and processed. For example, machine learning algorithms are used to mine hidden patterns and associations in the data to optimize the operation parameters of the boiler.

[0062] In this embodiment, based on the high-precision dynamic simulation and intelligent analysis capabilities of the boiler digital twin model, the system can predict the boiler wall temperature in real time and identify potential fault risks in advance. The boiler digital twin model ensures the accuracy and reliability of the prediction results through real-time update and dynamic adjustment.

[0063] In this solution, the boiler digital twin model can be dynamically updated according to real-time data to ensure that the model always reflects the actual operation state of the boiler. This real-time update mechanism enables the model to adapt to various changes during the operation of the boiler, thereby improving the response speed and prediction accuracy of the system.

[0064] Among them, the boiler digital twin model combines sensor data, historical operation data, and real-time operation parameters. Through multi-source data fusion technology, the accuracy and reliability of the model are improved. This fusion not only includes the simple superposition of data but also involves the in-depth integration and analysis of data.

[0065] Specifically, the boiler digital twin model provides decision-making support for operators through intelligent analysis and prediction. For example, when an abnormal wall temperature is predicted, the model can automatically recommend adjusting operation parameters such as the air volume of the burner or the feed water flow rate to optimize the operation of the boiler.

[0066] Among them, the prediction result is the wall temperature of the pipe.

[0067] In this solution, based on the prediction and early warning module 140, machine learning algorithms are used to predict the wall temperature, obtain the prediction result, and provide fault diagnosis and early warning functions according to the magnitude of the prediction result. By real-time monitoring the wall temperature data and combining machine learning algorithms, potential faults are identified and early warnings are issued.

[0068] Optionally, the prediction and early warning module 140 includes a wall temperature early warning unit; among them,

[0069] The wall temperature early warning unit is used to predict the wall temperature of the boiler digital twin model in the digital twin model module by using a deep learning model to obtain the prediction result.

[0070] Specifically, a trained deep learning model is used to perform real-time prediction on the boiler wall temperature. The model inputs include boiler operation parameters and wall coordinates, and the output is the wall temperature of the corresponding position.

[0071] Through the digital twin model combined with deep learning algorithms, high-precision real-time prediction of the boiler wall temperature is achieved. Compared with traditional numerical simulation methods, this system has a significant improvement in calculation speed and response time, and the calculation time is within 0.1 seconds. The system uses a deep neural network (DNN) model and combines a fast database expansion method to significantly improve the prediction accuracy of heat transfer deterioration problems. The mean square error (MSE) of the model is only 0.0053, and the accuracy rate (AUC5) reaches 0.988.

[0072] Optionally, the prediction and early warning module 140 further includes a fault diagnosis unit and an early warning push unit; among them,

[0073] The fault diagnosis unit is used to perform fault diagnosis according to the prediction result to determine the early warning information;

[0074] The early warning push unit is used to push the early warning information in real time.

[0075] Among them, the warning information includes the fault location, fault type, and recommended measures.

[0076] In this embodiment, by real-time monitoring of the wall temperature data and combining with machine learning algorithms, potential faults are identified and warning information is issued.

[0077] Furthermore, the warning information is pushed to the operator in real time through the interaction interface to help them adjust the boiler operation parameters in a timely manner and avoid accidents.

[0078] By real-time monitoring of the wall temperature data and combining with machine learning algorithms, potential faults are identified in advance and warnings are issued, improving the safety of boiler operation. The warning information can be pushed to the operator in real time to help them adjust the boiler operation parameters in a timely manner and avoid accidents.

[0079] Optionally, the system further includes an interaction interface module 150; among them,

[0080] The interaction interface module 150 is connected to the prediction and warning module 140 and is used to display the prediction results and warning information through the interaction interface.

[0081] In this solution, by providing an intuitive visual interface, the operator can monitor and adjust the operation parameters of the boiler in real time.

[0082] Optionally, the interaction interface module 150 includes a visual display unit, a real-time monitoring unit, and a historical data query unit; among them,

[0083] The visual display unit is used to display the prediction results and warning information through the interaction interface;

[0084] The real-time monitoring unit is used to monitor the operation status of the boiler digital twin model in real time through the interface;

[0085] The historical data query unit is used to query the historical operation data of the boiler digital twin model.

[0086] The operation parameters of the boiler are displayed through the interaction interface to help the operator adjust the boiler operation parameters in a timely manner and avoid accidents.

[0087] In this solution, through digital twin technology and machine learning algorithms, the system can automatically adjust the boiler operation parameters, reduce manual intervention, and improve the automation level. The system provides an intuitive visualization interface, through which operators can monitor the boiler operation status in real time, adjust the operation parameters, and the operation is simple. It reduces the operation cost and improves the economy. Through the real-time monitoring and early warning functions, the system can detect potential problems in advance, reduce the downtime and maintenance costs. The system can optimize the boiler operation parameters according to the real-time data, improve the combustion efficiency, and reduce the energy consumption. Based on the multi-physical field coupling modeling and deep learning theory, through theoretical derivation and model optimization, the scientificity and reliability of the system are ensured. Through verification with a large amount of experimental data, the prediction accuracy and response speed of the system under different working conditions meet the actual application requirements. It provides an intuitive visualization interface to display the boiler wall temperature distribution, prediction results, and early warning information. Operators can monitor the boiler operation status in real time and adjust the operation parameters through the interface. It supports querying and analyzing historical data to help operators better understand the boiler operation trend.

[0088] Furthermore, by introducing distributed computing and edge computing technologies, some computing tasks are assigned to the edge devices on the boiler site, reducing data transmission latency and improving the real-time performance and response speed of the system.

[0089] In this embodiment, the control circuit includes a data acquisition module 110, a data processing and analysis module 120, and a prediction and early warning module 140. Each module is connected through a high-speed communication interface to ensure real-time data transmission and processing.

[0090] In this solution, high-performance FPGA or DSP chips are used to achieve fast data processing and analysis. The sensor network uses high-precision sensors to ensure the accuracy of the data. The sensors are connected to the data acquisition module 110 through a signal conditioning circuit. The data acquisition module 110 is connected to the data processing and analysis module 120 through an Ethernet interface. The prediction and early warning module 140 is connected to the user interaction interface through a communication interface.

[0091] Furthermore, the boiler digital twin model obtains various types of data during the operation of the boiler in real time through the data acquisition module 110. These data include sensor data, real-time operation parameters, and historical operation data. The data acquisition module 110 serves as a bridge connecting the physical world and the digital twin system, ensuring that the digital twin model can reflect the actual operation status of the boiler in real time. The specific connection method is as follows: The data acquisition module 110 collects data in real time from high-precision temperature sensors, pressure sensors, flow sensors, etc. arranged on the boiler wall and transmits it to the boiler digital twin model. The data acquisition module 110 also collects the operation parameters adjusted by the operator in real time, such as the total coal volume, burner air volume, etc., and transmits these parameters to the boiler digital twin model. The data acquisition module 110 reads historical operation data from the database and integrates it into the boiler digital twin model for model training and optimization.

[0092] Among them, the boiler digital twin model interacts with the physical equipment of the boiler through the control circuit. The control circuit is responsible for converting the prediction results and optimization suggestions of the boiler digital twin model into specific control instructions and sending them to the boiler DCS system. The specific connection method is as follows: The boiler digital twin model generates optimization control instructions according to the prediction results, such as instructions to adjust the burner air volume or feed water flow. The control circuit sends these instructions to the DCS control system of the boiler (such as fans, water pumps, etc.) to achieve real-time adjustment of the boiler operation parameters.

[0093] In this embodiment, the digital twin model is connected to the SIS system to ensure real-time data transmission. The specific connection method is as follows: The boiler digital twin model receives the real-time data collected by the sensor through the network interface and sends the prediction results and control instructions to the DCS system through hard wiring. The SIS system feeds back the execution results to the boiler digital twin model to achieve closed-loop control.

[0094] Through digital twin technology and multi-source data fusion, combined with deep learning algorithms, this system not only improves the prediction accuracy and response speed but also has the functions of fault diagnosis and early warning, significantly improving the safety and economy of boiler operation.

[0095] The technical solution of the embodiment of the present invention realizes high-precision simulation and prediction of the boiler wall temperature by means of digital twin technology in combination with multi-source data during boiler operation. The digital twin model of the boiler is used for real-time dynamic update to ensure that the model can reflect the changes in the boiler wall temperature in real time and meet the requirements of real-time monitoring and prediction. On the basis of the digital twin model of the boiler, a fault diagnosis function is added. By real-time monitoring the boiler wall temperature data and combining machine learning algorithms, potential faults can be identified in advance. The system can automatically push deterioration warnings, extract and inform defects that will be caused by changes in a certain index, and avoid the occurrence of accidents in advance. Through digital twin technology, refined management of the boiler heating surface is realized, instant feedback on the equipment status is provided, and the maintenance and metal inspection and testing work of the boiler heating surface are guided. Through real-time monitoring and fault warning, the downtime and maintenance costs are reduced, and the service life of the equipment is extended. Digital twin technology is used to manage the entire life cycle of industrial boilers, providing a comprehensive solution from design, operation to maintenance.

[0096] Embodiment 2

[0097] Figure 2 The following is a flowchart of a method for predicting the boiler wall temperature based on a digital twin model provided by Embodiment 2 of the present invention. The method for predicting the boiler wall temperature based on a digital twin model is executed by a system for predicting the boiler wall temperature based on a digital twin model. As Figure 2 shown, the method includes:

[0098] S210. Collect the operation data of the boiler in real time; wherein, the operation data includes sensor data, historical operation data, and real-time operation parameters of the boiler.

[0099] In this solution, the sensor data includes temperature sensor data, pressure sensor data, flow sensor data, etc.

[0100] Among them, the historical operation data refers to the operation data of the boiler at historical moments. The historical operation data includes historical sensor data and historical real-time operation parameters of the boiler.

[0101] In this embodiment, during the operation of the boiler, the real-time operation parameters of the boiler mainly refer to the parameters that need to be monitored and adjusted in real time during the operation of the boiler. The real-time operation parameters of the boiler include combustion-related parameters, steam-related parameters, water-related parameters, and other operation parameters, etc. Among them, the combustion-related parameters include the total coal amount, the air volume of the burner, and the burnout air ratio. The total coal amount refers to the total amount of coal burned by the boiler per unit time, usually expressed in tons per hour (t / h). Adjusting the total coal amount in real time can control the combustion intensity and ensure that the output power of the boiler matches the load demand. The air volume of the burner refers to the air volume of each burner, with the unit of kilograms per second (kg / s). By adjusting the air volume of the burner, the combustion efficiency can be optimized and pollutant emissions can be reduced. The burnout air ratio refers to the air volume used to control the later stage of combustion to ensure that the fuel is fully burned and incomplete combustion losses are reduced. The steam-related parameters include steam pressure, steam temperature, and steam mass flow rate. The steam pressure refers to the steam pressure generated by the boiler, with the unit of megapascals (MPa). Monitoring the steam pressure in real time can ensure that the boiler operates within a safe pressure range. The steam temperature refers to the temperature of the steam, with the unit of degrees Celsius (°C) or Kelvin (K). By controlling the steam temperature, the efficiency of the steam turbine can be optimized. The steam mass flow rate refers to the mass flow rate of the steam passing through the boiler per unit time, with the unit of tons per hour (t / h). Adjusting the steam flow rate in real time can ensure that the output of the boiler matches the system demand. The water-related parameters include feed water flow rate, the water temperature at the outlet of the economizer, and the drum water level. The feed water flow rate refers to the amount of water entering the boiler per unit time, with the unit of tons per hour (t / h). Monitoring and adjusting the feed water flow rate in real time can ensure the stability of the boiler water level. The water temperature at the outlet of the economizer refers to the water temperature at the outlet of the economizer, with the unit of degrees Celsius (°C). By controlling the water temperature at the outlet of the economizer, the thermal efficiency of the boiler can be optimized. The drum water level refers to the water level height in the boiler drum, with the unit of millimeters (mm). Monitoring the drum water level in real time can prevent the boiler from drying out or overfilling. Other operation parameters include the flue gas temperature, the burner configuration mode, and the furnace negative pressure. The flue gas temperature refers to the temperature of the flue gas discharged from the boiler, with the unit of degrees Celsius (°C). By monitoring the flue gas temperature, the combustion efficiency and heat loss can be judged. The burner configuration mode refers to the operation mode of the burner, for example, which burners are in operation and which are in standby state. By adjusting the burner configuration, the combustion process can be optimized. The furnace negative pressure refers to the pressure inside the furnace, usually expressed in pascals (Pa). By controlling the furnace negative pressure, the stability and safety of the combustion process can be ensured.

[0102] In this solution, the sensors include temperature sensors, pressure sensors, and flow sensors. Each sensor is arranged on the boiler wall. The operating data of the boiler are collected in real time by each sensor.

[0103] S220. Preprocess the operating data to obtain target data.

[0104] In this solution, preprocessing mainly refers to the process of collecting, cleaning, transforming, etc. the original data to improve data quality, enhance the performance of the algorithm model, and make it more suitable for subsequent analysis, modeling and other tasks.

[0105] Specifically, preprocessing includes data cleaning, data denoising and data normalization. Data cleaning mainly refers to discovering and correcting identifiable errors in data files, including checking data consistency, handling invalid values and missing values, etc. Data denoising means removing noise from the collected data, restoring the true appearance of the data, improving data quality for subsequent analysis and processing. Data normalization is to map the data to a specific interval or range according to a certain method, so that the data has a unified scale and distribution, and eliminates the influence of dimension and order of magnitude between different features. By cleaning, denoising and normalizing the operation data, the data quality can be improved.

[0106] In this solution, machine learning algorithms can be used to extract the features related to the wall temperature from the preprocessed data, and the feature variables with high correlation degrees are screened out through the grey relational analysis method, and the feature variables are used as the target data. Among them, the target data includes combustion efficiency, flue gas temperature, working medium flow rate, etc.

[0107] S230. Construct a boiler digital twin model based on the said operation data and target data.

[0108] In this embodiment, a boiler digital twin model is constructed according to the operation data and target data, and the model is dynamically updated and simulated in combination with the data. Specifically, the digital twin technology can be used to construct a boiler digital twin model according to the operation data and target data. Digital twin is a simulation process that makes full use of data such as physical models, sensors, operation history, etc., integrates multiple disciplines, multiple physical quantities, multiple scales, and multiple probabilities, and completes the mapping in the virtual space, so as to reflect the whole life cycle process of the corresponding physical equipment.

[0109] Specifically, a boiler digital twin model is constructed based on the physical model, operation data and target data of the boiler. Among them, the boiler digital twin model includes the geometric structure, combustion process, heat transfer process, etc. of the boiler. Constructing a boiler digital twin model that includes multiple physical fields such as the geometric structure, combustion process, and heat transfer process of the boiler can comprehensively reflect the physical process of the boiler operation.

[0110] Furthermore, through the provided data, the boiler digital twin model is dynamically updated to ensure that the model can reflect the actual operation state of the boiler in real time. Combining sensor data, historical operation data and real-time operation parameters of the boiler improves the accuracy and reliability of the model.

[0111] S240. Use a machine learning algorithm to predict the wall temperature of the boiler digital twin model, obtain a prediction result, and perform fault diagnosis and early warning based on the prediction result.

[0112] Among them, the prediction result is the pipe wall temperature.

[0113] In this solution, a machine learning algorithm is used to predict the wall temperature, obtain a prediction result, and provide a fault diagnosis and early warning function based on the size of the prediction result. By real-time monitoring the wall temperature data and combining with the machine learning algorithm, potential faults are identified and early warnings are issued.

[0114] Specifically, a trained deep learning model is used to perform real-time prediction on the boiler wall temperature. The model inputs include boiler operation parameters and wall coordinates, and the output is the pipe wall temperature at the corresponding position.

[0115] Furthermore, the early warning information is pushed to the operator in real time through the interaction interface to help them adjust the boiler operation parameters in time and avoid accidents. Among them, the early warning information includes the fault location, fault type, and recommended measures.

[0116] By real-time monitoring the wall temperature data and combining with the machine learning algorithm, potential faults are identified and early warnings are issued in advance, improving the safety of boiler operation. The early warning information can be pushed to the operator in real time to help them adjust the boiler operation parameters in time and avoid accidents.

[0117] The technical solution of the embodiment of the present invention is to collect the operation data of the boiler in real time, preprocess the operation data to obtain target data, construct a boiler digital twin model according to the operation data and target data, use a machine learning algorithm to predict the wall temperature of the boiler digital twin model, obtain a prediction result, and perform fault diagnosis and early warning based on the prediction result. By implementing this technical solution, high-precision and real-time simulation and prediction of the boiler wall temperature are realized, improving the safety and economy of boiler operation.

[0118] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0119] The above specific implementation manners do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A boiler wall temperature prediction system based on a digital twin model, characterized in that: include: Data acquisition module, data processing and analysis module, digital twin model module and prediction and early warning module; among them, The data acquisition module is connected to the data processing and analysis module, and is used to collect the operation data of the boiler in real time, and transmit the operation data to the data processing and analysis module; wherein the operation data includes sensor data, historical operation data and real-time operation parameters of the boiler; The data and analysis module is connected to the digital twin model module, and is used to pre-process the operation data to obtain target data, and transmit the target data to the digital twin model module; The digital twin model module is used to build a digital twin model of the boiler according to the operating data and the target data; The prediction and early warning module is connected to the digital twin model module, and is used to predict the wall temperature of the boiler digital twin model in the digital twin model module by using a machine learning algorithm to obtain a prediction result, and perform fault diagnosis and early warning based on the prediction result.

2. The system according to claim 1, characterized in that The data acquisition module includes a sensor network unit, a data transmission unit and a data storage unit; wherein, The sensor network unit is used to collect the operation data of the boiler in real time based on the pre-arranged sensors; The data transmission unit is used to transmit the operating data to the data processing and analysis module via industrial Ethernet or wireless communication technology; The data storage unit is used to store the operation data in a distributed database.

3. The system according to claim 1, characterized in that The data processing and analysis module includes a data preprocessing unit and a feature extraction unit; wherein, The data preprocessing unit is used to clean, denoise and normalize the operation data to obtain preprocessed data; The feature extraction unit is used to extract features from the preprocessed data to obtain target data.

4. The system according to claim 1, characterized in that The data processing and analysis module also includes a model training unit; wherein, The model training unit is used to train the deep learning model to be trained according to the target data to obtain a deep learning model.

5. The system according to claim 1, characterized in that The digital twin model module includes a model building unit and a real-time updating unit; wherein, The model building unit is used to build a boiler digital twin model according to the operating data and the target data; The real-time updating unit is used to update the operating parameters of the boiler digital twin model in real time.

6. The system according to claim 1, characterized in that The prediction and warning module includes a wall temperature warning unit; wherein, The wall temperature early warning unit is used to predict the wall temperature of the boiler digital twin model in the digital twin model module by using a deep learning model to obtain a prediction result.

7. The system according to claim 1, characterized in that The prediction and warning module also includes a fault diagnosis unit and a warning push unit; wherein, The fault diagnosis unit is used to perform fault diagnosis according to the prediction result and determine the warning information; The warning push unit is used to push the warning information in real time.

8. The system according to claim 1, characterized in that The system also includes an interactive interface module; wherein, The interactive interface module is connected to the prediction and warning module and is used to display the prediction results and warning information through the interactive interface.

9. The system according to claim 8, characterized in that The interactive interface module includes a visual display unit, a real-time monitoring unit and a historical data query unit; wherein, The visualization display unit is used to display the prediction results and warning information through an interactive interface; The real-time monitoring unit is used to monitor the operating status of the boiler digital twin model in real time through an interface; The historical data query unit is used to query the historical operation data of the boiler digital twin model.

10. A boiler wall temperature prediction method based on a digital twin model, characterized in that: include: Collecting boiler operation data in real time; wherein the operation data includes sensor data, historical operation data and real-time boiler operation parameters; Preprocessing the operating data to obtain target data; constructing a digital twin model of the boiler based on the operating data and the target data; The wall temperature of the boiler digital twin model is predicted by using a machine learning algorithm to obtain a prediction result, and fault diagnosis and early warning are performed based on the prediction result.

Images