Combustor life cycle energy efficiency management platform based on digital twinning
The burner full life cycle energy efficiency management platform built with digital twin technology solves the problems of traditional platforms being unable to adaptively adjust and identify deep-level energy efficiency degradation. It realizes in-depth expansion and early identification of burner full life cycle energy efficiency assessment, and improves the accuracy and stability of energy efficiency management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HEHE ENERGY (BEIJING) CO LTD
- Filing Date
- 2025-11-28
- Publication Date
- 2026-07-31
AI Technical Summary
Existing burner lifecycle energy efficiency management platforms fail to assess the energy depreciation process from the perspective of entropy production. Traditional digital twin models cannot adaptively adjust, leading to a gradual mismatch between the model and the physical entity, and an inability to deeply identify energy efficiency degradation.
A burner lifecycle energy efficiency management platform based on digital twins is adopted. Multi-scale entropy generation data is acquired through real-time sensing units, digital twin construction units are coupled in two channels, monitoring units build a long-term database, collaborative optimization units perform simulation optimization, and energy efficiency performance is recorded through trajectory writing units, thereby realizing multi-dimensional energy efficiency assessment and adaptive adjustment.
It has achieved a deeper expansion of the full-cycle energy efficiency assessment of burners, enabling early identification of deep-seated energy efficiency degradation, improving the accuracy and stability of energy efficiency management, reducing energy consumption, and enhancing economic benefits.
Smart Images
Figure CN121580842B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of burner energy efficiency data management technology, specifically to a burner lifecycle energy efficiency management platform based on digital twins. Background Technology
[0002] A burner is a device that mixes fuel (such as natural gas, fuel oil, coal, etc.) with air or oxygen and then burns it to generate heat or power. It is widely used in industrial boilers, heating furnaces, internal combustion engines, gas turbines, and other fields, and is one of the key pieces of equipment for energy conversion and utilization. Burner design must consider factors such as fuel type, combustion efficiency, emission control, and safety to ensure a highly efficient, clean, and stable combustion process.
[0003] Throughout the burner's entire lifecycle, advanced sensor technology, data analysis methods, and intelligent control strategies are employed to achieve lifecycle energy efficiency management of the combustion process. For example, IoT technology is used to collect burner operating data, enabling comprehensive and systematic management and optimization of its energy utilization efficiency. This process aims to ensure that the burner maintains efficient and stable operation at different stages, thereby reducing energy consumption, emissions, and improving economic benefits.
[0004] However, most current burner lifecycle energy efficiency management platforms are mainly based on thermal efficiency calculations, with a single dimension of energy efficiency assessment. They fail to assess the energy depreciation process from the perspective of entropy production and cannot identify deep-seated energy efficiency degradation. At the same time, the parameters of traditional digital twin models are fixed and cannot be adaptively adjusted with changes in the burner, resulting in a gradual mismatch between the model and the actual system. Summary of the Invention
[0005] To address these issues, this invention provides a burner lifecycle energy efficiency management platform based on digital twins.
[0006] To achieve the above objectives, the present invention provides the following technical solution:
[0007] The burner full life cycle energy efficiency management platform based on digital twins includes a real-time sensing unit, a digital twin construction unit, a monitoring unit, a collaborative optimization unit, and a trajectory writing unit.
[0008] The real-time sensing unit synchronously collects heat flux density and temperature gradient data within the burner based on a sensor array, and then calculates the comprehensive entropy production index generated by heat conduction and viscous dissipation in real time; the comprehensive entropy production index... The calculation formula is as follows:
[0009]
[0010] in, This serves as a reference value for the entropy production rate. This is a reference value for spectral entropy. and All are weighted coefficients. For entropy production rate, Spectral entropy ;
[0011] The digital twin construction unit uses parallel physical twin channels and data twin channels to construct the digital twin. The two channels are bidirectionally coupled through a parameter interaction layer. The physical twin channel provides training constraints for the neural network that conform to physical laws, while the data twin channel dynamically adjusts its boundary conditions and internal parameter settings, ultimately outputting complete flow field information.
[0012] The monitoring unit continuously collects the pressure difference between the inlet and outlet of the burner through a differential pressure transmitter, and constructs a long-term operating database to store the differential pressure readings at different time points and the corresponding entropy production spatial characteristics, forming status data.
[0013] The collaborative optimization unit receives complete flow field information from the digital twin construction unit and status data from the monitoring unit. Through the virtual environment constructed by the digital twin, and using an objective function, it optimizes the burner's comprehensive entropy production index over a future period. Integral minimization;
[0014] The trajectory writing unit can build a distributed storage architecture based on blockchain technology, record the entropy production rate and comprehensive entropy production index of the burner at different load points under standard operating conditions, and set normal operating range thresholds for each monitoring point to form a traceable initial energy efficiency performance file.
[0015] Furthermore, the real-time sensing unit includes an entropy production decomposition subunit, an online analysis subunit, and a data fusion subunit;
[0016] The entropy production decomposition subunit adopts a hierarchical computing architecture: at the macroscopic scale, pressure gradient and velocity gradient are obtained through sensor arrays; at the microscopic scale, viscosity matrix under the current operating conditions is obtained by calling the fuel characteristic database.
[0017] The online analysis subunit is built based on a spectral entropy feature analyzer, which can calculate spectral entropy for the characteristic spectral lines of OH and CH radicals. This establishes a correlation between spectral entropy and combustion stability; spectral entropy... The calculation formula is as follows:
[0018]
[0019] in, For the first A range of amplitude values, For the first The probability of a range of amplitude values occurring;
[0020] The data fusion subunit can use a feature fusion method to perform spatiotemporal matching of local entropy yield and spectral entropy, and calculate a comprehensive entropy yield index. .
[0021] Furthermore, the physical twin channel is used to analyze the fundamental physical phenomena in the combustion process and construct a digital mapping environment.
[0022] Furthermore, the data twin channel is constructed using a long short-term memory neural network, with its input layer receiving real-time operating parameters and its hidden layer capturing the dynamic temporal characteristics of the combustion process through a gating mechanism.
[0023] Furthermore, the objective function is as follows:
[0024]
[0025] in, , and All are preset coefficients. This represents the entropy production rate.
[0026] Furthermore, the monitoring unit can also input status data into the digital twin construction unit to perform targeted reconstruction of the structured grid system in the physical twin channel, so that the spatial structure of the digital twin is consistent with the actual form of the physical entity.
[0027] Furthermore, the simultaneous trajectory writing unit also includes a data analysis engine, which can extract trends from the historical entropy production data of the burner and calculate the slope k of the trend term in the dataset through a sliding window.
[0028] Furthermore, the formula for calculating the slope k is as follows:
[0029]
[0030] in, For the first The data collection time at each point in time. The average time across all time points. For the first spectral entropy at each time point , For all spectral entropy The average value, This represents the total number of monitoring sessions.
[0031] This invention has the following advantages: Through multi-dimensional entropy production analysis and an adaptive digital twin architecture, it achieves a deep extension of burner full-cycle energy efficiency assessment. The multi-scale entropy flow data provided by the real-time sensing unit establishes a precise input foundation for the digital twin construction unit, enabling the virtual model to comprehensively reflect the irreversible characteristics of the burner during the combustion process from the perspective of entropy production.
[0032] Meanwhile, the digital twin building unit deeply integrates physical laws and operational data through a dual-channel coupling mechanism, realizing a panoramic digital mapping of the combustion process. This provides a high-fidelity simulation environment for the collaborative optimization unit and significantly improves the early identification capability of deep-level energy efficiency degradation.
[0033] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. Attached Figure Description
[0034] To more intuitively illustrate the prior art and this application, exemplary drawings are provided below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing this application; for example, based on the technical concept disclosed in this application and the exemplary drawings, those skilled in the art are able to easily make conventional adjustments or further optimizations to the addition / reduction / classification, specific shapes, positional relationships, connection methods, size ratios, etc. of certain units (components).
[0035] Figure 1 This is an architecture diagram of the burner full life cycle energy efficiency management platform based on digital twins of this invention. Detailed Implementation
[0036] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these embodiments are merely for further explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Technical engineers in the field can make some non-essential improvements and adjustments to the present invention based on the above-described content. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0037] Please see Figure 1 The burner full life cycle energy efficiency management platform based on digital twins includes: a real-time sensing unit, a digital twin construction unit, a monitoring unit, a collaborative optimization unit, and a trajectory writing unit.
[0038] The real-time sensing unit synchronously collects heat flux density and temperature gradient data within the burner based on a sensor array. It then calculates the comprehensive entropy production index generated by heat conduction and viscous dissipation in real time, converting the temperature and velocity field gradients into a scalar field of entropy production rate. This transforms traditional temperature and pressure signals into a microscopic entropy production index that directly characterizes energy quality loss.
[0039] The real-time sensing unit also includes an entropy production decomposition subunit, an online analysis subunit, and a data fusion subunit. The entropy production decomposition subunit adopts a hierarchical computing architecture: at the macroscopic scale, it obtains pressure and velocity gradients through a sensor array; at the microscopic scale, it retrieves the viscosity matrix under the current operating conditions from a fuel characteristic database. This provides multi-scale, multi-physics entropy flow data input for subsequent digital twins.
[0040] The sensor array includes a miniature infrared thermal imaging array and a temperature sensor array. When collecting temperature gradient data, the miniature infrared thermal imaging array (pixel resolution ≥ 640×512) captures the dynamic distribution of the temperature field in real time at a sampling frequency of 50Hz.
[0041] Real-time calculation of the entropy production rate of each spatial node due to heat conduction and viscous dissipation. This forms a three-dimensional entropy production rate distribution cloud map, showing the entropy production rate. The calculation formula is as follows:
[0042]
[0043] Where q is the heat flux density. Let T be the rate of temperature change, and T be the local temperature. For dynamic viscosity, Let be the velocity gradient tensor.
[0044] The online analysis subunit is built upon a spectral entropy feature analyzer, which uses a fiber optic spectral detection system to acquire the ultraviolet-visible emission spectrum of the flame. Spectral entropy is calculated based on the characteristic spectral lines of OH and CH radicals. This establishes a correlation between spectral entropy and combustion stability. Spectral entropy is calculated. The calculation formula is as follows:
[0045]
[0046] Among them, spectral intensity The amplitude range is divided into equal parts. For the first A range of amplitude values, For the first The probability of a given amplitude range occurring is such that the more uniformly the value is distributed, the greater the entropy value, indicating an increase in the degree of disorder in the combustion process; if it is concentrated in a specific interval (such as the stable combustion stage), the entropy value approaches zero.
[0047] The data fusion subunit receives real-time data streams from two subunits and uses a feature fusion method to spatiotemporally match the local entropy yield and spectral entropy to calculate the comprehensive entropy yield index. This index simultaneously reflects the irreversible nature of the combustion process in terms of microstructure, macroscopic system, and spectral characteristics, enabling a multi-dimensional and unified quantitative evaluation of energy loss during combustion. (Comprehensive Entropy Production Index) The calculation formula is as follows:
[0048]
[0049] in, This serves as a reference value for the entropy production rate. This is a reference value for spectral entropy. and All are weighted coefficients, and + =1.
[0050] The digital twin construction unit uses parallel physical twin channels and data twin channels to construct the digital twin. The two channels are bidirectionally coupled through a parameter interaction layer. The physical twin channel provides training constraints for the neural network that conform to physical laws, while the data twin channel dynamically adjusts its boundary conditions and internal parameter settings, ultimately outputting complete flow field information, providing a high-fidelity virtual environment for subsequent optimization.
[0051] The physical twin channel is used to analyze the fundamental physical phenomena in the combustion process and construct a digital mapping environment. This channel first divides the combustion chamber into a structured grid system, spatially discretizing fluid motion and heat exchange. Simultaneously, based on the principles of mass conservation, momentum balance, and energy conversion, it tracks the velocity changes, temperature evolution, and component concentration distribution of fluid micro-elements within the burner in real time. By loading fuel injection parameters, airflow conditions, and wall heat exchange characteristics, it generates complete flow field information reflecting the current operating conditions.
[0052] The data twin channel is constructed using a long short-term memory neural network. Its input layer receives real-time operating parameters, its hidden layer captures the dynamic temporal characteristics of the combustion process through a gating mechanism, and its output layer predicts key state parameters. The two twins are bidirectionally coupled through a parameter interaction layer: the physical twin channel inputs complete flow field information as a physical guidance signal into the neural network's training process, providing constraints that conform to physical laws for the digital twin; simultaneously, the neural network feeds back the correction amounts predicted based on actual operating data to the physical twin channel, dynamically adjusting its boundary conditions. The two channels exchange data in real time through the parameter interaction layer.
[0053] The monitoring unit continuously collects the pressure difference between the burner inlet and outlet via a differential pressure transmitter and builds a long-term operational database, storing differential pressure readings at different time points and their corresponding entropy production spatial characteristics, forming traceable status data. Simultaneously, it identifies trends in the accumulated data. When a sustained increase in differential pressure is detected and a high-entropy production region stably appears at a specific location in the flow channel, it determines that structural changes may be occurring in that region. By comparing the current entropy production distribution with the baseline state, and considering the magnitude of differential pressure changes, the monitoring unit locates the possible areas of geometric deformation and their severity.
[0054] The monitoring unit inputs state data into the digital twin construction unit, and performs targeted reconstruction of the structured mesh system in the physical twin channel. At the identified geometrically deformed and refined mesh nodes, the boundary geometry description is adjusted in the deformed regions to ensure that the spatial structure of the digital twin matches the actual shape of the physical entity. After adjustment, the new mesh is immediately activated for state simulation, achieving synchronous mapping at the geometric level.
[0055] The collaborative optimization unit receives complete flow field information from the digital twin construction unit and status data from the monitoring unit. Through the virtual environment constructed by the digital twin, it performs forward-looking simulation and strategy verification of the combustion process, and uses an objective function to evaluate the comprehensive entropy production index of the burner over a future period. Integral minimization is employed, using a digital twin to simulate the entropy production trend, generating an optimal control sequence, and distributing it to the actuators. This achieves real-time matching between the combustion process and entropy production characteristics. The objective function is as follows:
[0056]
[0057] in, , and These are all preset coefficients that can be dynamically adjusted according to the current combustion strategy.
[0058] After each control cycle is completed, the actual sensor data (such as the newly generated temperature field distribution) is read and compared with the predicted value to calculate the residual. If the residual exceeds the threshold (such as temperature deviation > ±15℃), the Bayesian inference algorithm is activated to update the weighting coefficients of the digital twin, and the control sequence for the next cycle is regenerated.
[0059] The trajectory writing unit can build a distributed storage architecture based on blockchain technology, record the entropy production rate and comprehensive entropy production index of the burner at different load points under standard operating conditions, and set normal operating range thresholds for each monitoring point to form a traceable initial energy efficiency performance file.
[0060] During operation, the trajectory writing unit continuously receives output data from each unit and uses a time series database to compress and store the monitoring data to obtain a dataset, while retaining the timestamp, operating condition label, and quality control identifier of the original data.
[0061] The trajectory writing unit also includes a data analysis engine, capable of extracting trends from historical entropy production data of the burner and calculating the slope k of the trend term in the dataset using a sliding window. The formula for calculating the slope k is as follows:
[0062]
[0063] in, For the first The data collection time at each point in time. The average time across all time points. For the first spectral entropy at each time point , For all spectral entropy The average value, This represents the total number of monitoring sessions.
[0064] When the slope of change remains positive and exceeds a preset threshold for multiple consecutive monitoring cycles, the system determines that the burner has entered an accelerated degradation stage, automatically generates an energy efficiency decline warning, and locates the specific component (such as the nozzle or heat exchange surface). This mechanism enables accurate identification of hidden performance degradation processes from massive amounts of operational data, providing a quantitative basis for preventive maintenance.
[0065] This invention achieves a deeper expansion of burner full-cycle energy efficiency assessment through multi-dimensional entropy production analysis and an adaptive digital twin architecture. The multi-scale entropy flow data provided by the real-time sensing unit establishes a precise input foundation for the digital twin building unit, enabling the virtual model to comprehensively reflect the irreversible characteristics of the burner during combustion from an entropy production perspective. The digital twin building unit, through a dual-channel coupling mechanism, deeply integrates physical laws with operational data, achieving a panoramic digital mapping of the combustion process. This provides a high-fidelity simulation environment for the collaborative optimization unit, significantly improving the early identification capability of deep-seated energy efficiency degradation.
[0066] Meanwhile, the collaborative work of the monitoring unit and the trajectory writing unit constructs a complete burner lifecycle management system. The structural deformation data identified by the monitoring unit drives the real-time mesh reconstruction of the digital twin, ensuring the consistency between the virtual model and the physical entity; the full lifecycle entropy data established by the trajectory writing unit not only records the decision trajectory of the collaborative optimization unit, but also identifies performance degradation trends through the data analysis engine, providing historical basis for the optimization unit's strategy adjustment.
[0067] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A burner lifecycle energy efficiency management platform based on digital twins, characterized in that, It includes a real-time sensing unit, a digital twin construction unit, a monitoring unit, a collaborative optimization unit, and a trajectory writing unit; The real-time sensing unit synchronously collects heat flux density and temperature gradient data within the burner based on a sensor array, and then calculates the comprehensive entropy production index generated by heat conduction and viscous dissipation in real time; the comprehensive entropy production index... The calculation formula is as follows: ， in, This serves as a reference value for the entropy production rate. This is a reference value for spectral entropy. and All are weighted coefficients. For entropy production rate, Spectral entropy; The real-time sensing unit includes an entropy generation decomposition subunit, an online analysis subunit, and a data fusion subunit; The entropy production decomposition subunit adopts a hierarchical computing architecture: at the macroscopic scale, pressure gradient and velocity gradient are obtained through sensor array; at the microscopic scale, viscosity matrix under the current operating condition is obtained by calling fuel characteristic database. The online analysis subunit is built based on a spectral entropy feature analyzer, which can calculate spectral entropy for the characteristic spectral lines of OH and CH radicals. This establishes a correlation between spectral entropy and combustion stability; The data fusion subunit can use feature fusion methods to spatiotemporally match local entropy production rates and spectral entropy to calculate a comprehensive entropy production index. ; The digital twin construction unit uses parallel physical twin channels and data twin channels to construct the digital twin. The two channels are bidirectionally coupled through a parameter interaction layer. The physical twin channel provides training constraints for the neural network that conform to physical laws, while the data twin channel dynamically adjusts its boundary conditions and internal parameter settings, ultimately outputting complete flow field information. The monitoring unit continuously collects the pressure difference between the inlet and outlet of the burner through a differential pressure transmitter, and constructs a long-term operating database to store the differential pressure readings at different time points and the corresponding entropy production spatial characteristics, forming status data. The collaborative optimization unit receives complete flow field information from the digital twin construction unit and status data from the monitoring unit. Through the virtual environment constructed by the digital twin, and using an objective function, it optimizes the burner's comprehensive entropy production index over a future period. Integral minimization is performed; the objective function is as follows: ， in, , and All are preset coefficients. The rate of entropy production; The trajectory writing unit can construct a distributed storage architecture based on blockchain technology, record the entropy production rate and comprehensive entropy production index of the burner at different load points under standard operating conditions, and set normal operating range thresholds for each monitoring point to form a traceable initial energy efficiency performance file. The trajectory writing unit also includes a data analysis engine, which can extract trends from historical entropy production data of the burner and calculate the slope k of the trend term in the dataset through a sliding window.
2. The burner lifecycle energy efficiency management platform based on digital twins according to claim 1, characterized in that, The spectral entropy The calculation formula is as follows: ， in, For the first A range of amplitude values, For the first The probability of a range of amplitudes occurring.
3. The burner lifecycle energy efficiency management platform based on digital twins according to claim 1, characterized in that, The physical twin channel is used to analyze the fundamental physical phenomena in the combustion process and construct a digital mapping environment.
4. The burner lifecycle energy efficiency management platform based on digital twins according to claim 1, characterized in that, The data twin channel is constructed using a long short-term memory neural network. Its input layer receives real-time operating parameters, and its hidden layer captures the dynamic temporal characteristics of the combustion process through a gating mechanism.
5. The burner lifecycle energy efficiency management platform based on digital twins according to claim 1, characterized in that, The monitoring unit can also input status data into the digital twin construction unit to perform targeted reconstruction of the structured grid system in the physical twin channel, so that the spatial structure of the digital twin is consistent with the actual form of the physical entity.
6. The burner lifecycle energy efficiency management platform based on digital twins according to claim 1, characterized in that, The formula for calculating the slope k is as follows: ， in, For the first The data collection time at each point in time. The average time across all time points. For the first spectral entropy at each time point , For all spectral entropy The average value, This represents the total number of monitoring sessions.