Multi-temperature-zone independent temperature control electric heater and heating method

Abstract

The application discloses a multi-temperature-zone independent temperature control electric heater and a heating method, belongs to the technical field of temperature control, and can be used in an industrial control system. The application provides a technical scheme for obtaining multi-dimensional data and establishing a dynamic parameter model, combining equipment state evaluation and a multi-objective optimization algorithm, realizing multi-temperature-zone independent precise temperature control and energy consumption optimization, and having the advantages of improving heating efficiency, reducing energy consumption and guaranteeing equipment reliability.

Description

Technical Field

[0001] This application relates to the field of temperature control technology, and in particular to a multi-temperature zone independently temperature-controlled electric heater and a heating method. Background Technology

[0002] In the petrochemical industry, heating oil in pipelines or storage tanks is a crucial step in ensuring its fluidity and processing efficiency. Traditional oil heating methods typically employ zone heating, maintaining the oil at a suitable temperature by controlling the temperature of different sections.

[0003] However, several problems exist in the relevant technologies: First, there is a lack of comprehensive perception of petroleum physical properties, heater status and environmental conditions, resulting in insufficient temperature control accuracy; second, there is thermal interference between different temperature zones, making it difficult to achieve truly independent and precise temperature control; and third, the ability to optimize energy consumption and coordinate equipment control is weak, making it impossible to simultaneously achieve temperature stability, optimal energy efficiency and equipment reliability.

[0004] These issues directly affect the economy and safety of the petroleum heating process, and there is an urgent need for a comprehensive solution that can achieve independent and precise temperature control in multiple temperature zones while optimizing energy consumption and equipment operating status. Summary of the Invention

[0005] This application provides a multi-temperature zone independently controlled electric heater and heating method, which can achieve independent and precise temperature control in multiple temperature zones while optimizing energy consumption and equipment operating status. The technical solution is as follows:

[0006] On one hand, a heating method is provided, the method comprising:

[0007] The system acquires petroleum heating monitoring data, operating data of the multi-zone independently temperature-controlled electric heater, environmental data, and operating condition data. Petroleum heating monitoring data includes real-time temperature distribution data of multiple heating zones collected by distributed temperature sensors on the electric heater, petroleum fluid characteristic data collected by the fluid characteristic detection module, and system energy consumption data collected by the energy consumption metering unit. Petroleum fluid characteristic data includes crude oil composition data, physical property data, and flow characteristic data. Crude oil composition data includes water content, wax content, and asphaltenes content; physical property data includes density, specific heat capacity, and thermal conductivity; flow characteristic data includes viscosity, pour point, and freezing point; system energy consumption data includes heating power, electrical energy consumption, and thermal efficiency; and environmental data includes ambient temperature and humidity. The system uses operating data, including pipeline pressure data and real-time oil flow data. Based on oil heating monitoring data, environmental data, electric heater operation data, and historical heater maintenance data, it determines the characteristic parameters of oil heating status and the characterization parameters of electric heater operation status. Based on these parameters, the system determines temperature control commands, energy consumption optimization commands, and heater coordination commands. Temperature control commands control the electric heating elements in each heating section, energy consumption optimization commands control the energy distribution strategy of the electric heater, and coordination commands coordinate the operation of multiple heaters. The system executes these commands to perform independent temperature control operations on each heating section through which the oil flows.

[0008] Furthermore, this application proposes determining the characteristic parameters of oil heating status and the characterization parameters of electric heater operation status based on oil heating monitoring data, environmental data, electric heater operation data, and historical heater maintenance data. This includes: determining the characteristic parameters of heating status using a multimodal data fusion algorithm based on oil heating monitoring data and environmental data, wherein the multimodal data fusion algorithm adopts an adaptive weighted fusion mechanism and dynamically adjusts the fusion weights according to the data quality of each sensor; and determining the characterization parameters of heater operation status using a heater status assessment model based on electric heater operation data and historical heater maintenance data.

[0009] Furthermore, this application proposes to determine heating status characteristic parameters based on petroleum heating monitoring data and environmental data using a multimodal data fusion algorithm. This includes: spatiotemporally aligning real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data from the petroleum heating monitoring data to generate a spatiotemporally synchronized multi-source data sequence; dynamically determining the confidence weight of each data source based on the multi-source data sequence using an adaptive weighted fusion mechanism, with the weight of real-time temperature distribution data determined based on its measurement accuracy and real-time performance, the weight of petroleum fluid characteristic data determined based on the detection cycle and reliability, and the weight of environmental data determined based on its correlation with the heating process; and using a weighted fusion algorithm to extract and fuse features from the weighted multi-source data sequence to obtain heating status characteristic parameters, including temperature field distribution characteristics, heat flux density characteristics, and energy efficiency characteristics.

[0010] Furthermore, this application proposes to perform spatiotemporal alignment on real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data from petroleum heating monitoring data to generate a spatiotemporally synchronized multi-source data sequence. This includes: preprocessing the real-time temperature distribution data to extract temperature measurement values ​​for each heating segment at the same timestamp and labeling the corresponding spatial location information; performing time interpolation on the petroleum fluid characteristic data to align the sampling period with the acquisition time sequence of the real-time temperature distribution data; performing time synchronization on the system energy consumption data and environmental data to ensure that the timestamps are consistent with the timestamps of the real-time temperature distribution data; and combining the time-synchronized and spatially labeled real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data according to the time sequence to obtain a spatiotemporally synchronized multi-source data sequence.

[0011] Furthermore, this application proposes that the electric heater operating data includes heater operating parameters, energy consumption data, and system status data; the heater historical maintenance data includes maintenance records, fault history, and component replacement records; based on the electric heater operating data and heater historical maintenance data, the heater operating status characterization parameters are determined through a heater status assessment model, including: performing a correlation analysis on the heater operating parameters, energy consumption data, and system status data with maintenance records, fault history, and component replacement records to obtain the correlation analysis results; based on the correlation analysis results and the target matching degree, the heater operating status characterization parameters are determined through the heater status assessment model, where the target matching degree is the matching degree between the heater operating parameters and the fault modes in the fault history; the heater operating status characterization parameters include heater operating efficiency indicators, heater energy consumption characteristic parameters, and heater health status assessment data.

[0012] Furthermore, this application proposes to conduct a correlation analysis on heater operating parameters, energy consumption data, and system status data with maintenance records, fault history, and component replacement records to obtain correlation analysis results. This includes: extracting time-series data on load rate, operating temperature, and runtime from heater operating parameters; instantaneous power and cumulative energy consumption from energy consumption data; and heater start-up / shutdown frequency and operational stability indicators from system status data; dynamically matching runtime time-series data, cumulative energy consumption time-series data, and system status data with maintenance cycles and maintenance items from maintenance records, operating parameter characteristics and fault types before and after faults from fault history, and component lifespan and performance changes before and after replacement from component replacement records to obtain corresponding operating mode similarity and correlation coefficients. System status data includes heater start-up / shutdown frequency and operational stability indicators. Based on the operating mode similarity and correlation coefficients, correlation analysis results are generated. These results describe the correlation between the current operating status of the electric heater and historical maintenance, fault, and component replacement behaviors. Finally, based on the correlation analysis results and target matching... The heater condition assessment model determines the heater's operating status characteristics, including: inputting correlation analysis results and target matching degree into the heater condition assessment model; weighting and fusing the correlation analysis results and target matching degree through the heater condition assessment model, with the weight of the correlation analysis results determined based on their correlation with the current operating conditions, and the weight of the target matching degree determined based on the severity and frequency of historical fault patterns; determining the heater's comprehensive health score and remaining life prediction value based on the fused characteristics through the heater condition assessment model, with the comprehensive health score characterizing the overall performance status of the heater and the remaining life prediction value estimating the heater's working time; and generating heater operating efficiency indicators, heater energy consumption characteristic parameters, and heater health status assessment data based on the comprehensive health score and remaining life prediction value through the heater condition assessment model, including heater operating efficiency indicators (heating efficiency coefficient and power conversion efficiency), heater energy consumption characteristic parameters (energy consumption rate per unit time and energy consumption trend prediction value), and heater health status assessment data (fault warning level and maintenance urgency indicators).

[0013] Furthermore, this application proposes determining temperature control commands, energy consumption optimization commands, and heater coordination commands based on heating status characteristic parameters, operating condition data, and heater operating status characterization parameters. This includes: determining petroleum rheological characteristic parameters, including apparent viscosity, flow index, and stability coefficient, using a petroleum fluid thermodynamic analysis model based on the heating status characteristic parameters and real-time petroleum flow data in the operating condition data; and generating temperature control commands, energy consumption optimization commands, and heater coordination commands using a multi-objective collaborative optimization algorithm based on the heater operating status characterization parameters and petroleum rheological characteristic parameters.

[0014] Furthermore, this application proposes determining petroleum rheological property parameters based on heating state characteristic parameters and real-time petroleum flow data in operating condition data through a petroleum fluid thermodynamic analysis model. This includes: inputting temperature field distribution characteristics, heat flux density characteristics, and real-time petroleum flow data from the heating state characteristic parameters into the petroleum fluid thermodynamic analysis model. The petroleum fluid thermodynamic analysis model is a simulation model constructed based on non-Newtonian fluid mechanics principles, including a viscosity-temperature characteristic determination module and a rheological behavior prediction module. Through the viscosity-temperature characteristic determination module, the apparent viscosity variation curve of petroleum in different temperature zones is determined based on the temperature field distribution characteristics and petroleum composition data. Through the rheological behavior prediction module, petroleum rheological property parameters are determined based on the apparent viscosity variation curve, heat flux density characteristics, and real-time petroleum flow data. These petroleum rheological property parameters include apparent viscosity, flow index, and stability coefficient. The petroleum rheological property parameters are used to characterize the flow properties and stability of petroleum under current heating conditions.

[0015] Furthermore, this application proposes to generate temperature control commands, energy consumption optimization commands, and heater coordination commands based on heater operating status characterization parameters and petroleum rheological characteristic parameters through a multi-objective collaborative optimization algorithm. This includes: determining the temperature control priority and power allocation constraints for each heating zone based on equipment health status assessment data in the heater operating status characterization parameters and stability coefficients in the petroleum rheological characteristic parameters; employing a constrained reinforcement learning algorithm to simultaneously optimize temperature control stability, energy consumption efficiency, and equipment operating reliability to generate preliminary control commands, which include target temperatures for each temperature zone, power allocation ratios, and equipment coordination timing; performing conflict detection and boundary condition verification on the preliminary control commands, including verifying whether thermal interference between temperature setpoints of each temperature zone exceeds the allowable range, and verifying whether power allocation exceeds the heater's maximum load capacity and whether the temperature rise rate is within the equipment safety limits; and outputting the preliminary control commands as temperature control commands, energy consumption optimization commands, and heater coordination commands after the preliminary control commands pass conflict detection and boundary condition verification.

[0016] Furthermore, this application proposes to execute temperature control commands, energy consumption optimization commands, and heater coordination commands to perform independent temperature control operations on multiple heating sections through which the oil flows, including: adjusting the output power of the electric heating elements in each heating section according to the temperature control command to control the temperature of each temperature zone to reach the target temperature value; adjusting the power distribution strategy of the electric heaters according to the energy consumption optimization command to shut down or reduce the heating power of non-critical temperature zones; and coordinating the working sequence between multiple heaters according to the heater coordination command to avoid the impact on the power supply caused by the simultaneous start-up and shutdown of multiple heaters.

[0017] As can be seen from the above, the oil multi-temperature zone independent temperature control heating method and system provided in this application achieves independent and precise temperature control and energy consumption optimization in multiple temperature zones by acquiring multi-dimensional data and establishing a dynamic parameter model, combined with equipment status assessment and multi-objective optimization algorithm. It has the advantages of improving heating efficiency, reducing energy consumption and ensuring equipment reliability.

[0018] On one hand, an electric heater is provided, the electric heater including one or more processors and one or more memories, the one or more memories storing at least one computer program, the computer program being loaded and executed by the one or more processors to implement the heating method.

[0019] On one hand, a computer-readable storage medium is provided, wherein at least one computer program is stored in the computer-readable storage medium, the computer program being loaded and executed by a processor to implement the heating method.

[0020] On one hand, a computer program product or computer program is provided, which includes program code stored in a computer-readable storage medium. The processor of the electric heater reads the program code from the computer-readable storage medium and executes the program code, causing the electric heater to perform the heating method described above. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the implementation environment of a heating method provided in an embodiment of this application;

[0023] Figure 2 This is a flowchart of a heating method provided in an embodiment of this application;

[0024] Figure 3 This is a flowchart of another heating method provided in an embodiment of this application;

[0025] Figure 4 This is a flowchart of another heating method provided in the embodiments of this application;

[0026] Figure 5 This is a flowchart of another heating method provided in the embodiments of this application;

[0027] Figure 6 This is a schematic diagram of the structure of an electric heater provided in an embodiment of this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0029] In this application, the terms "first," "second," etc., are used to distinguish identical or similar items with essentially the same function. It should be understood that there is no logical or temporal dependency between "first," "second," and "nth," nor are there any restrictions on quantity or execution order.

[0030] Artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain better results.

[0031] Normalization: Mapping sequences of values ​​with different ranges to the interval (0, 1) to facilitate data processing. In some cases, normalized values ​​can be directly expressed as probabilities.

[0032] Embedded coding, mathematically speaking, represents a correspondence, that is, mapping data in space X to space Y using a function F. This function F is injective, and the mapping result preserves the structure. An injective function means that the mapped data uniquely corresponds to the original data, and preserving the structure means that the size relationship between the original and mapped data is the same. For example, if there are data X1 and X2 before mapping, after mapping we get Y1 corresponding to X1 and Y2 corresponding to X2. If the original data X1 > X2, then correspondingly, the mapped data Y1 > Y2. For words, this means mapping words to another space to facilitate subsequent machine learning and processing.

[0033] Attention weights represent the importance of a piece of data during training or prediction. Importance indicates the magnitude of the influence of input data on output data. Data with high importance corresponds to higher attention weights, while data with low importance corresponds to lower attention weights. The importance of data varies in different scenarios, and training the model to assign attention weights is essentially the process of determining data importance.

[0034] Figure 1 This is a schematic diagram illustrating the implementation environment of a heating method provided in this application embodiment. See also... Figure 1The implementation environment may include an electric heater 100 with independent temperature control in multiple temperature zones. The electric heater 100 includes a heating component 110 and a controller 140. In this embodiment of the application, the heater 100 is a general term for the heating system. The controller 140 is used to control the heating component 110 to realize the independent temperature control capability of the electric heater 100 in multiple temperature zones.

[0035] In related technologies, petroleum heating processes commonly employ zoned temperature control. However, due to a lack of comprehensive sensing capabilities regarding petroleum physical properties, equipment status, and environmental conditions, temperature control precision is insufficient. Thermal interference exists between different temperature zones, making independent and precise temperature control difficult. Traditional methods have significant shortcomings in energy consumption optimization and equipment collaborative control, failing to balance temperature stability and equipment reliability. For example, in high-viscosity crude oil transportation scenarios, existing heating systems struggle to detect the impact of crude oil composition changes on heat conduction in real time, leading to excessive local temperature fluctuations and causing pipeline waxing or energy waste.

[0036] To address the aforementioned issues, the inventors identified three core contradictions in the relevant technologies: isolated multi-source data leading to insufficient basis for control decisions; lack of dynamic suppression methods for thermal interference in temperature zones; and difficulty in simultaneously achieving equipment coordination and energy efficiency optimization. Analysis revealed that the coupling relationship between petroleum fluid characteristics and heater operating status was not fully explored, and the impact of equipment health status on power distribution was not considered during control command generation. Based on this, the inventors proposed establishing a multi-dimensional data sensing system, constructing a dynamic control model by integrating petroleum physical properties, environmental parameters, and equipment operating data. Furthermore, heater health status is incorporated into the optimization objective, achieving synergy between temperature control and equipment protection.

[0037] Therefore, this application proposes a heating method, see [link to relevant documentation]. Figure 2 Taking the controller as the executing entity as an example, the method includes the following steps:

[0038] 201. Acquire monitoring data for oil heating, operating data for electric heaters with independent temperature control in multiple temperature zones, environmental data, and operating condition data;

[0039] 202. Based on oil heating monitoring data, environmental data, electric heater operation data, and historical heater maintenance data, determine the characteristic parameters of oil heating status and the characterization parameters of electric heater operation status.

[0040] 203. Based on heating status characteristic parameters, operating condition data, and heater operating status characterization parameters, determine temperature control commands, energy consumption optimization commands, and heater coordination commands;

[0041] 204. Execute temperature control commands, energy consumption optimization commands, and heater coordination commands to perform independent temperature control operations on multiple heating sections through which the oil flows.

[0042] The petroleum heating monitoring data includes real-time temperature distribution data of multiple heating sections collected by distributed temperature sensors in the electric heater, petroleum fluid characteristic data collected by the fluid characteristic detection module, and system energy consumption data collected by the energy consumption metering unit. The petroleum fluid characteristic data includes crude oil composition data, physical property data, and flow characteristic data. Crude oil composition data includes water content, wax content, and asphaltene content; physical property data includes density, specific heat capacity, and thermal conductivity; flow characteristic data includes viscosity, pour point, and freezing point; system energy consumption data includes heating power, electrical energy consumption, and thermal efficiency; environmental data includes ambient temperature and humidity; and operating condition data includes pipeline pressure data and real-time petroleum flow rate data. The petroleum heating monitoring data refers to the multi-dimensional parameters of the heating process collected in real time through a distributed sensor network. Specifically, this can be achieved using an array of temperature sensors arranged along the pipeline axis, an online fluid analyzer, and smart meters to construct a comprehensive data view of the heating process. Environmental data includes external variables affecting heat exchange efficiency, which can be collected by temperature and humidity sensors placed around the heater to compensate for the impact of environmental factors on temperature control accuracy. Heating status characteristic parameters are comprehensive evaluation indicators generated through data fusion. Specifically, a spatiotemporal alignment algorithm can be used to integrate temperature field distribution and fluid characteristic data to characterize the heating state and energy transfer efficiency of petroleum. Heater coordination commands are timing control strategies that coordinate the operation of multiple devices. Specifically, a dynamic priority scheduling algorithm can be used to generate these commands to avoid power surges caused by simultaneous start-up and shutdown of multiple heaters.

[0043] Specifically, this method first acquires temperature distribution data for each heating zone through a distributed sensor network, and then monitors changes in crude oil composition in real time using an online fluid analyzer. Environmental sensors simultaneously collect external temperature and humidity data, forming a complete set of heating environment parameters. A spatiotemporal alignment algorithm is used to synchronize fluid characteristic data and temperature data at different sampling frequencies, eliminating data delay errors. Historical heater maintenance data and real-time operating parameters are correlated and analyzed to assess the weight of equipment health status on power allocation. During the control command generation phase, petroleum rheological parameters and equipment health indicators are input into the optimization model to dynamically adjust the target temperature and power allocation ratio for each temperature zone. In the execution phase, a time-sharing start-up strategy is used to coordinate the operation of multiple heaters, reducing power load fluctuations through peak-shifting operation.

[0044] Compared to related technologies, traditional methods use a fixed weighting strategy to process sensor data, failing to consider the impact of dynamic changes in data quality on the fusion results. This solution uses an adaptive weighting mechanism to adjust data confidence in real time. For example, when a temperature sensor in a certain section drifts, its weight is automatically reduced while the proportion of data from adjacent sensors is increased. Existing temperature control systems mostly use independent PID controllers; this solution uses a fluid thermodynamics model to predict changes in the rheological properties of petroleum and adjusts the heating power allocation in advance. Traditional equipment collaborative control only considers load balancing; this solution incorporates the predicted remaining lifespan of the equipment into the optimization objective, extending the service life of key components while ensuring temperature control accuracy.

[0045] Through the above technical solutions, this application can effectively suppress heat conduction interference within temperature ranges. For example, it can quickly adjust the power of the upstream heating zone when the wax content suddenly increases to prevent wax deposition from spreading downstream. A dynamic power allocation strategy reduces energy consumption in non-critical sections while maintaining the target temperature; for example, it automatically reduces the heating power of the insulation section when the ambient temperature rises. A coordinated equipment control strategy avoids instantaneous overload caused by the simultaneous start-up of multiple heaters; for example, it staggers the start-up sequence of heaters in different sections when pipeline flow fluctuates, ensuring stable power supply operation.

[0046] This application further proposes the following technical solutions, see [link / reference] Figure 3 Taking the controller as the executing entity as an example, the following steps are included.

[0047] 301. Based on oil heating monitoring data and environmental data, heating status characteristic parameters are determined through a multimodal data fusion algorithm. The multimodal data fusion algorithm adopts an adaptive weighted fusion mechanism, which dynamically adjusts the fusion weights according to the data quality of each sensor.

[0048] 302. Based on the electric heater's operating data and historical maintenance data, the heater's operating status characterization parameters are determined through a heater status assessment model.

[0049] Among them, the multimodal data fusion algorithm refers to the method of spatiotemporal alignment and weight allocation of heterogeneous data collected by different types of sensors. Specifically, it can be implemented using time interpolation, spatial labeling, and dynamic confidence level determination to eliminate the impact of differences in data acquisition cycles and spatial locations on the fusion results. The adaptive weighted fusion mechanism dynamically adjusts the contribution weight of the data source in the fusion process based on the real-time quality of the data source. Specifically, it can be implemented through measurement accuracy assessment, real-time performance scoring, and reliability indicators to address the problem that static weight allocation cannot adapt to dynamic data changes. The heater condition assessment model is an assessment system built based on the correlation analysis of equipment operating parameters and historical maintenance records. Specifically, it can be implemented using dynamic time matching algorithms and correlation coefficient determination to identify potential correlations between the current operating status of the equipment and historical fault patterns.

[0050] Specifically, a multimodal data fusion algorithm is used to perform spatiotemporal synchronization processing on real-time temperature distribution data, petroleum fluid characteristic data, and environmental data, eliminating timestamp discrepancies and spatial location differences between different data sources. An adaptive weighted fusion mechanism is employed to dynamically assign weights to the synchronized multi-source data sequences; for example, real-time temperature data is given a higher weight for real-time performance, while petroleum component data is given a higher weight for reliability. The weighted fusion algorithm extracts key characteristic parameters such as temperature field distribution and heat flux density to form a comprehensive index reflecting the current heating status. Simultaneously, a heater condition assessment model is used to perform correlation analysis between operating parameters and maintenance records. For example, the current load rate is matched with load fluctuation patterns prior to the occurrence of failures to determine equipment health scores and remaining life predictions, generating quantitative parameters characterizing the equipment's operating status.

[0051] Compared to related technologies, traditional methods typically use fixed weights to fuse multi-source data, which cannot adapt to dynamic changes in sensor data quality, resulting in insufficient accuracy of the fusion results. This solution, however, effectively addresses the impact of fluctuating temperature sensor accuracy and differences in petroleum characteristic detection cycles by dynamically adjusting weights. Furthermore, existing equipment condition assessments are largely limited to real-time operating parameter analysis, neglecting the correlation between historical maintenance data and failure modes. This solution, by establishing a correlation analysis model for full lifecycle data, can more accurately predict potential equipment failure risks.

[0052] Through the above technical solution, this application achieves accurate fusion of multi-dimensional data on the petroleum heating process, which to some extent solves the problem of control command deviation caused by data quality fluctuations in traditional methods. Simultaneously, by integrating real-time equipment operating data with historical maintenance experience, the reliability of equipment health status assessment is improved, providing accurate equipment status parameters to support subsequent optimized control and avoiding energy waste and operational risks caused by equipment performance degradation.

[0053] This application further proposes to perform spatiotemporal alignment of real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data in petroleum heating monitoring data to generate a spatiotemporally synchronized multi-source data sequence. Based on the multi-source data sequence, an adaptive weighted fusion mechanism is used to dynamically determine the confidence weight of each data source. The weight of real-time temperature distribution data is determined based on its measurement accuracy and real-time performance, the weight of petroleum fluid characteristic data is determined based on the detection cycle and reliability, and the weight of environmental data is determined based on its correlation with the heating process. A weighted fusion algorithm is used to extract and fuse features from the weighted multi-source data sequence to obtain heating status characteristic parameters, including temperature field distribution characteristics, heat flux density characteristics, and energy efficiency characteristics.

[0054] Spatiotemporal alignment refers to synchronizing data from different timestamps and spatial locations. This can be achieved using time interpolation, spatial labeling, and time series matching methods to ensure consistency across multiple data sources in both time and space. Adaptive weighted fusion mechanisms dynamically adjust weights based on the measurement accuracy, real-time performance, reliability, and relevance to the heating process of the data sources. This can be implemented using confidence assessment algorithms and dynamic weight allocation models to optimize the accuracy of data fusion. Weighted fusion algorithms extract and integrate features from the weighted multi-source data. This can be achieved using principal component analysis, feature concatenation, or deep learning models to generate feature parameters that comprehensively reflect the state of the heating system.

[0055] Specifically, in the spatiotemporal alignment process, real-time temperature distribution data is aligned spatially by extracting temperature measurements at the same timestamp and labeling them with spatial location information. Petroleum fluid characteristic data undergoes time interpolation to ensure its sampling period matches the acquisition time series of the real-time temperature data. System energy consumption data and environmental data are time-synchronized to maintain timestamp consistency with the real-time temperature data. The resulting multi-source data sequence is completely synchronized in time and space, providing a foundation for subsequent fusion analysis. In dynamic weight determination, real-time temperature distribution data is assigned a higher weight due to its high measurement accuracy and real-time performance. Petroleum fluid characteristic data is weighted based on the detection cycle length and sensor reliability, while environmental data is weighted according to its impact on the heating process. Through a weighted fusion algorithm, temperature field distribution characteristics, heat flux density characteristics, and energy efficiency characteristics are extracted and fused to form comprehensive heating situation characteristic parameters.

[0056] Compared to related technologies, traditional methods typically use fixed weights or single-dimensional data for fusion, failing to consider issues such as spatiotemporal data asynchrony and unreasonable weight allocation, leading to biases in feature parameters. This solution eliminates data inconsistencies through spatiotemporal alignment and adaptively adjusts the contribution of data sources using a dynamic weight allocation mechanism, thus more accurately reflecting the true state of the heating system. For example, in related technologies, the long detection cycle of petroleum fluid characteristic data results in a mismatch between the timestamps of real-time temperature data and the fusion results in a lag; this solution avoids such errors by using time interpolation to achieve data synchronization.

[0057] Through the above technical solutions, this application has to some extent solved the problem of inaccurate heating status characteristic parameters caused by spatiotemporal asynchrony and unreasonable weight allocation of multi-source data. By spatiotemporal alignment and dynamic weight allocation, the fusion accuracy of multi-dimensional data is effectively improved, making the extraction of temperature field distribution, heat flux density, and energy efficiency characteristics more accurate, thereby enhancing the ability to perceive the operating status of the heating system.

[0058] This application further proposes preprocessing the real-time temperature distribution data, extracting the temperature measurement values ​​of each heating section at the same timestamp and labeling the corresponding spatial location information; performing time interpolation processing on the petroleum fluid characteristic data to align the sampling period with the acquisition time series of the real-time temperature distribution data; performing time synchronization processing on the system energy consumption data and environmental data to ensure that the timestamps are consistent with the real-time temperature distribution data; and combining the time-synchronized and spatially labeled real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data according to the time series to generate a spatiotemporally synchronized multi-source data sequence.

[0059] The preprocessing of real-time temperature distribution data involves extracting temperature values ​​with the same time signature from the raw data acquired by distributed temperature sensors. This can be achieved using a timestamp matching algorithm combined with sensor spatial coordinate mapping to eliminate data misalignment caused by time deviations between different sensors. Temporal interpolation processing of petroleum fluid characteristic data involves converting discrete property detection data into a continuous time series. This can be achieved using linear interpolation or spline interpolation algorithms to address the periodic differences between low-frequency sampling of fluid characteristics and high-frequency temperature monitoring. Time synchronization processing of system energy consumption data involves unifying the clock references of different acquisition systems to the time axis of the main control system. This can be achieved using network time protocol synchronization or timestamp rewriting mechanisms to eliminate data timing errors caused by clock asynchrony among multiple systems. Spatial location information labeling involves establishing a mapping relationship between temperature measurements and the physical location of the heating section. This can be achieved by matching sensor numbers with a heating section topology database to establish a spatial coordinate system for the temperature field distribution.

[0060] Specifically, in the data preprocessing stage, the clock calibration module of the distributed temperature sensor first extracts the temperature measurements of each heating section under the same time reference, and simultaneously generates three-dimensional spatial annotation data by associating the sensor installation location coordinates. For petroleum fluid characteristic data, a continuous data sequence consistent with the temperature acquisition frequency is generated using a cubic spline interpolation algorithm based on the time interval of its discrete sampling points. The system energy consumption data and environmental monitoring data are forcibly aligned to the time reference of the temperature data through a timestamp rewriting module. Finally, the spatiotemporally aligned multi-source data are arranged in chronological order to form a multi-dimensional data matrix containing spatial coordinates, timestamps, temperature values, physical property parameters, energy consumption indicators, and environmental parameters, providing spatiotemporally consistent data input for subsequent data fusion.

[0061] Compared to related technologies, traditional methods often employ a single time reference or ignore spatial location correlations, leading to spatiotemporal data misalignment in temperature field analysis. For example, related technologies fail to label the spatial locations of distributed sensors, resulting in an inaccurate reflection of temperature gradient distribution between heating zones; the lack of interpolation processing for low-frequency property detection data causes a mismatch between the time series of temperature and property data; and the use of independent clocks by different systems leads to time deviations between energy consumption data and environmental parameters. This solution, however, establishes a dual spatiotemporal alignment mechanism, ensuring both the spatial topological accuracy of temperature data and precise time synchronization of multi-source data, effectively solving the spatiotemporal matching problem of cross-system, multi-frequency data to a certain extent.

[0062] Through the above technical solution, this application can eliminate the spatiotemporal mismatch problem caused by differences in acquisition cycles, clock asynchrony, and spatial location of multi-source data, ensuring that the input data required for subsequent multimodal data fusion maintains strict consistency in time and space dimensions. This improves the accuracy of determining heating state characteristic parameters, provides a reliable spatiotemporal correlation data foundation for generating temperature control commands, and avoids control errors caused by data misalignment.

[0063] This application further proposes to conduct a correlation analysis on heater operating parameters, energy consumption data, and system status data with maintenance records, fault history, and component replacement records to obtain correlation analysis results. Based on the correlation analysis results and the target matching degree, the heater operating status characterization parameters are determined through a heater status assessment model. The target matching degree is the matching degree between the heater operating parameters and the fault modes in the fault history. The heater operating status characterization parameters include heater operating efficiency indicators, heater energy consumption characteristic parameters, and heater health status assessment data.

[0064] Among these, correlation analysis refers to the process of dynamically matching time-series data such as equipment runtime and cumulative energy consumption with maintenance cycles and fault characteristics. This can be achieved using dynamic time warping algorithms or similarity measurement algorithms to discover potential correlations between equipment operating modes and historical maintenance behaviors. Target matching degree is a quantitative indicator of the similarity between current operating parameters and historical fault characteristics. This can be achieved using pattern recognition algorithms or similarity determination models to assess the closeness of the current operating state to known fault modes. The heater condition assessment model is a deterministic model that integrates multi-dimensional data to assess equipment health. This can be achieved by combining weighted fusion algorithms with machine learning models to generate a comprehensive health score and remaining life prediction value.

[0065] Specifically, during implementation, the following steps are taken: First, time-series data on load rate, operating temperature, and runtime from the heater's operating parameters, and time-series data on instantaneous power and cumulative energy consumption from the energy consumption data. These data are then time-series matched with maintenance cycles and items in the maintenance records, and simultaneously compared with the operating parameter characteristics before and after faults in the fault history. A dynamic time-matching algorithm is used to determine the similarity of operating patterns, generating a correlation coefficient to characterize the relationship between the current operating state and historical maintenance behaviors. Subsequently, the correlation analysis results and the target matching degree are input into the heater condition assessment model, which performs weighted fusion based on the relevance weight of the current operating condition and the severity weight of the fault history patterns. The fused features are used to determine the comprehensive health score and remaining life prediction value, ultimately outputting multi-dimensional condition parameters including heating efficiency coefficient, energy consumption rate per unit time, and maintenance urgency indicators.

[0066] Compared to related technologies, traditional methods typically assess equipment status based solely on current operating parameters, neglecting the correlation between historical maintenance data and failure modes. Related technologies commonly employ single-parameter threshold judgments, such as relying solely on temperature or power exceeding limits for fault warnings, failing to identify the dynamic correlation between potential failure modes and operating parameters. This solution, by establishing a dynamic matching mechanism between operating data and maintenance records, can capture changes in the similarity between operating parameters and historical fault characteristics during equipment performance degradation, thereby identifying latent fault risks in advance.

[0067] Through the above technical solution, this application addresses to some extent the reliability judgment bias caused by the single data dimension in traditional equipment evaluation. By dynamically correlating and analyzing real-time operating data with historical maintenance records, the correlation between equipment performance degradation trends and potential failure modes can be accurately identified. The multi-dimensional state parameters generated based on the weighted fusion mechanism provide a comprehensive decision-making basis for formulating preventative maintenance strategies, including equipment health, energy efficiency, and maintenance urgency, effectively avoiding sudden equipment failures due to untimely maintenance.

[0068] This application further proposes to perform correlation analysis on heater operating parameters, energy consumption data, and system status data with maintenance records, fault history, and component replacement records to obtain correlation analysis results; extract time-series data of load rate, operating temperature, and running time from heater operating parameters, instantaneous power and cumulative energy consumption from energy consumption data, and heater start-up / shutdown frequency and operational stability indicators from system status data; and compare the running time-series data, cumulative energy consumption time-series data, and system status data with maintenance cycle and maintenance items from maintenance records, operating parameter characteristics and fault types before and after faults from fault history, and component lifespan and replacement records from component replacement records. The performance change data is dynamically matched over time to obtain the corresponding similarity and correlation coefficients of the operating modes; based on the similarity and correlation coefficients of the operating modes, correlation analysis results are generated; the correlation analysis results and target matching degree are input into the heater condition assessment model; the correlation analysis results and target matching degree are weighted and fused through the heater condition assessment model; based on the fused features, the heater's comprehensive health score and remaining life prediction value are determined through the heater condition assessment model; based on the comprehensive health score and remaining life prediction value, the heater operating efficiency index, heater energy consumption characteristic parameters, and heater health status assessment data are generated through the heater condition assessment model.

[0069] Dynamic time matching refers to aligning and analyzing the time series of real-time operational data with the time axis of historical maintenance events. This can be achieved using dynamic time warping algorithms to discover potential correlations between the current operating mode and historical fault events. Operating mode similarity refers to determining the degree of morphological similarity between the current operating parameter curve and the fault precursor curve. This can be achieved using dynamic time warping distance algorithms to quantify the degree to which the current state deviates from the normal pattern. The correlation coefficient is an indicator characterizing the statistical correlation between different data dimensions. This can be achieved using the Pearson correlation coefficient method to identify the correlation between component performance degradation and abnormal energy consumption. Weighted fusion involves assigning different weights based on the importance and reliability of the data source. This can be achieved using entropy weighting combined with expert experience rules to improve the accuracy of the condition assessment results.

[0070] Specifically, by extracting time series data of dynamic parameters such as load rate and operating temperature and matching them with maintenance cycles in maintenance records, abnormal states of overdue maintenance can be identified. Correlation analysis between cumulative energy consumption data and performance change data in component replacement records reveals the causal relationship between sudden increases in energy consumption and heating element aging. Similarity determination between operational stability indicators and operational parameter characteristics in fault history records can capture the correlation pattern between frequent relay starts and stops and contactor burnout faults. During model processing, when the current load rate is detected to exceed the historical normal range, the matching weight of similar fault patterns is automatically increased, focusing the evaluation model on high-risk fault types. By integrating health scores and remaining life prediction values, a maintenance urgency index is generated; when the health score falls below a preset threshold and the remaining life is close to the maintenance cycle, a secondary warning signal is triggered.

[0071] Compared to related technologies, traditional methods rely solely on isolated analysis of current operational data, failing to establish correlations with historical maintenance events and thus hindering the prediction of component lifespan degradation trends. This solution utilizes dynamic time-matching technology to correlate real-time data streams with a maintenance record database, identifying implicit correlations between heating element aging and energy consumption fluctuations. Existing evaluation models employ fixed weight allocation mechanisms, which struggle to adapt to varying data reliability under different operating conditions. This solution employs adaptive weight adjustments based on operational conditions, ensuring that health scores more closely reflect the actual state of the equipment.

[0072] Through the above technical solutions, this application achieves deep correlation analysis between the operating status of the electric heater and historical maintenance data, effectively identifying the potential link between equipment performance degradation and historical failure patterns. By dynamically adjusting the weight allocation mechanism of the evaluation model, the accuracy of health status assessment under different operating conditions is improved. The generated maintenance urgency index provides a quantitative basis for formulating preventive maintenance plans, avoiding unplanned downtime caused by sudden failures. The predicted energy consumption trend can guide the optimization of power allocation strategies, reducing ineffective energy consumption while ensuring heating stability.

[0073] This application further proposes the following technical solutions to determine the relevant instructions, see [link to relevant documentation]. Figure 4 Taking the controller as the executing entity as an example, the following steps are included.

[0074] 401. Based on the heating state characteristic parameters and real-time oil flow data in the operating condition data, the oil rheological characteristic parameters are determined by the oil fluid thermodynamic analysis model. The oil rheological characteristic parameters include apparent viscosity, flow index and stability coefficient.

[0075] 402. Based on the heater operating status characterization parameters and petroleum rheological characteristic parameters, a multi-objective collaborative optimization algorithm is used to generate temperature control commands, energy consumption optimization commands, and heater coordination commands.

[0076] The petroleum fluid thermodynamic analysis model refers to a simulation model built based on non-Newtonian fluid dynamics principles. Specifically, it can be implemented using a viscosity-temperature characteristic determination module and a rheological behavior prediction module. This module determines the apparent viscosity change curve based on temperature field distribution characteristics and petroleum component data, and predicts flow behavior by combining heat flux density characteristics and real-time petroleum flow data. The multi-objective collaborative optimization algorithm refers to a constrained reinforcement learning algorithm, specifically implemented using dynamic priority allocation and boundary condition verification mechanisms. This algorithm is used to simultaneously optimize temperature control stability, energy efficiency, and equipment operational reliability. Conflict detection and boundary condition verification refers to verifying the thermal interference range and equipment load capacity of initial control commands. This can be achieved through thermodynamic simulation and comparison of real-time monitoring data, ensuring that the commands meet process requirements and equipment safety specifications.

[0077] Specifically, the petroleum fluid thermodynamic analysis model first receives the temperature field distribution characteristics and heat flux density characteristics from the heating state characteristic parameters. Combined with real-time petroleum flow data, it generates an apparent viscosity change curve through a viscosity-temperature characteristic determination module. Based on this curve and real-time flow data, the rheological behavior prediction module outputs petroleum rheological characteristic parameters, including the flow index and stability coefficient. Subsequently, a multi-objective collaborative optimization algorithm dynamically generates temperature control priorities and power allocation constraints based on equipment health status assessment data from the heater operating status characterization parameters and the stability coefficient from the petroleum rheological characteristic parameters. A reinforcement learning algorithm simultaneously optimizes the temperature setpoint, power allocation ratio, and equipment coordination timing under these constraints, generating preliminary control commands. After thermal interference range verification, equipment load capacity verification, and temperature rise rate safety limit detection, these commands are finally output as executable temperature control commands, energy consumption optimization commands, and heater coordination commands.

[0078] Compared to related technologies, traditional methods typically rely solely on single temperature feedback for control, failing to consider the dynamic coupling relationship between petroleum rheological properties and equipment health status, resulting in a lack of global optimization in control commands. This solution, however, quantifies fluid characteristics through a petroleum fluid thermodynamic analysis model and combines it with a multi-objective collaborative optimization algorithm to achieve coordinated adjustment of equipment status and process parameters. This ensures that control commands simultaneously meet the requirements for precise temperature control, energy consumption optimization, and equipment health maintenance.

[0079] Through the above technical solutions, this application has to some extent solved the problem of multi-parameter coupling control in petroleum heating processes, achieving improved temperature control accuracy, optimized energy efficiency, and enhanced equipment synergy. Specifically, the dynamic temperature setpoint based on rheological characteristic analysis can avoid flow stagnation caused by sudden viscosity changes; the constrained reinforcement learning algorithm achieves optimal energy allocation while ensuring equipment safety; and the conflict detection mechanism effectively prevents thermal interference in multiple temperature zones and the risk of equipment overload.

[0080] This application further proposes a method to determine petroleum rheological parameters based on heating condition characteristic parameters and real-time petroleum flow data from operating conditions, using a petroleum fluid thermodynamic analysis model. This includes inputting the temperature field distribution characteristics, heat flux density characteristics, and real-time petroleum flow data from the heating condition characteristic parameters into the petroleum fluid thermodynamic analysis model. The model is a simulation model built based on non-Newtonian fluid mechanics principles, including a viscosity-temperature characteristic determination module and a rheological behavior prediction module. The viscosity-temperature characteristic determination module determines the apparent viscosity variation curves of petroleum in different temperature zones based on the temperature field distribution characteristics and petroleum composition data. The rheological behavior prediction module determines the petroleum rheological parameters based on the apparent viscosity variation curves, heat flux density characteristics, and real-time petroleum flow data. These parameters include apparent viscosity, flow index, and stability coefficient. The petroleum rheological parameters characterize the flow properties and stability of petroleum under current heating conditions.

[0081] The petroleum fluid thermodynamic analysis model is a simulation model based on non-Newtonian fluid mechanics principles. Specifically, it can be implemented using finite element analysis combined with rheological constitutive equations to simulate the viscosity changes and flow behavior of petroleum during heating. The viscosity-temperature characteristic determination module determines viscosity changes based on temperature field distribution characteristics and petroleum composition data. Specifically, it can be implemented using a temperature-viscosity relationship model based on the Arrhenius equation to establish a nonlinear relationship between temperature gradient and petroleum viscosity. The rheological behavior prediction module predicts flow characteristics based on viscosity change curves and real-time flow data. Specifically, it can be implemented using the Herschel-Bulkley model combined with the momentum conservation equation to determine the flow index and stability coefficient. Temperature field distribution characteristics refer to the internal temperature gradient data of petroleum obtained through distributed temperature sensors. Specifically, this can be acquired using thermocouple arrays or fiber optic temperature measurement systems to reflect temperature differences at different spatial locations during heating. Heat flux density characteristics refer to the heat transferred per unit area. Specifically, this can be determined using a heat flux meter combined with heater power data to characterize the energy transfer efficiency between the heater and the petroleum.

[0082] Specifically, the temperature field distribution characteristics are determined by real-time acquisition of temperature gradient data for each heating zone using distributed sensors. This data, combined with physical properties such as asphaltene content from petroleum component data, is input into the viscosity-temperature characteristic determination module. This module is based on a non-Newtonian fluid temperature-viscosity nonlinear relationship model, for example, using a modified Arrhenius equation, to determine the apparent viscosity change curves corresponding to different temperature zones. The rheological behavior prediction module receives viscosity curves, heat flux density characteristics, and real-time flow data. It establishes a numerical model coupling the momentum conservation equation and the rheological constitutive equation, for example, using the Herschel-Bulkley model to describe the relationship between shear stress and shear rate. Combined with heat flux density data, it determines the impact of energy dissipation on the flow, ultimately outputting a flow index characterizing flow performance and a coefficient reflecting flow stability. During this process, real-time flow data is used to dynamically correct flow boundary conditions, ensuring that the prediction results match actual operating conditions.

[0083] Compared to related technologies, traditional methods typically determine viscosity using a single temperature point or average temperature, failing to consider the spatial variability of fluid viscosity caused by temperature field distribution. This leads to inaccurate viscosity predictions for non-Newtonian fluids such as high-wax crude oil. Furthermore, rheological parameter determination in related technologies is largely based on static physical property data, neglecting the impact of real-time heat flux density changes on flow behavior, making it difficult to accurately reflect the dynamic rheological characteristics during heating. In addition, traditional models do not jointly analyze temperature gradient data with petroleum component data, failing to distinguish the differentiated effects of different crude oil components on viscosity-temperature characteristics.

[0084] Through the above technical solution, this application can accurately predict the rheological properties of non-Newtonian fluids under dynamic heating conditions, eliminate viscosity determination errors caused by uneven temperature field distribution, and solve the problem of inaccurate flow behavior prediction caused by the failure to consider dynamic changes in heat flux density and flow rate in traditional methods. By coupling temperature gradient data and component data, differentiated modeling of different crude oil types can be achieved, improving the accuracy of rheological parameter prediction and providing an accurate physical property parameter basis for multi-temperature zone coordinated control.

[0085] This application further proposes a method for generating temperature control commands, energy consumption optimization commands, and heater coordination commands based on heater operating status characterization parameters and petroleum rheological characteristic parameters using a multi-objective collaborative optimization algorithm. This includes: determining the temperature control priority and power allocation constraints for each heating zone based on equipment health status assessment data from the heater operating status characterization parameters and stability coefficients from the petroleum rheological characteristic parameters; employing a constrained reinforcement learning algorithm to simultaneously optimize temperature control stability, energy consumption efficiency, and equipment operating reliability to generate preliminary control commands, which include target temperatures for each temperature zone, power allocation ratios, and equipment coordination timing; performing conflict detection and boundary condition verification on the preliminary control commands, including verifying whether thermal interference between temperature setpoints of each temperature zone exceeds the allowable range, and verifying whether the power allocation exceeds the heater's maximum load capacity and whether the temperature rise rate is within the equipment safety limits; and outputting the preliminary control commands as temperature control commands, energy consumption optimization commands, and heater coordination commands after passing conflict detection and boundary condition verification.

[0086] Among these, equipment health status assessment data refers to quantitative indicators reflecting the current operational reliability of the electric heater. Specifically, it can be generated using a status assessment model based on historical maintenance records and real-time operating parameters, used to dynamically adjust the power allocation upper limit for each heating zone. The stability coefficient is a dynamic parameter characterizing the influence of temperature changes on the flow properties of oil. Specifically, it can be obtained by determining the correlation between oil viscosity and temperature gradient through a thermodynamic analysis model, used to determine the minimum temperature threshold for different temperature zones. Constrained reinforcement learning algorithms are optimization methods that use equipment operational reliability as a hard constraint. Specifically, they can be implemented using a Markov decision process framework combined with the Lagrange multiplier method, used to simultaneously optimize multiple conflicting objective functions. Conflict detection is the process of identifying heat conduction interference between temperature setpoints in multiple temperature zones. Specifically, it can be used to determine the temperature superposition effect of adjacent temperature zones using the heat conduction equation, used to avoid local overheating or temperature fluctuations. Boundary condition verification is the step of verifying whether control commands comply with the physical limitations of the equipment. Specifically, it can be achieved by querying the heater's rated power parameters and the safe threshold for the temperature rise rate, used to prevent equipment overload or over-temperature operation.

[0087] Specifically, when determining temperature control priorities, heating sections with higher maintenance urgency indicators in the equipment health status assessment data are assigned lower power limits to prevent equipment failure; simultaneously, regions with lower stability coefficients in the petroleum rheological properties parameters are set with higher temperature limits to maintain flow performance. A constrained reinforcement learning algorithm constructs a multi-objective reward function incorporating temperature deviation, energy consumption indicators, and equipment health, generating a set of candidate instructions that satisfy power allocation constraints in each iteration. The conflict detection module solves the partial differential equation of heat conduction to predict the temperature distribution change in adjacent temperature zones after the execution of candidate instructions; when the change exceeds a preset threshold, instruction correction is triggered. The boundary condition verification module compares the power allocation ratio in the candidate instructions with the heater nameplate parameters in real time; if it detects that the power demand of a certain section exceeds 85% of the maximum load capacity, it automatically lowers the target temperature setpoint for that section.

[0088] In some specific implementations, the constrained reinforcement learning algorithm can employ a deep deterministic policy gradient framework, where the action space is limited to a temperature setting range that satisfies equipment health constraints. Conflict detection can utilize finite element simulation tools to predetermine the thermal interference matrix under typical operating conditions, and quickly determine the feasibility of commands through table lookup. Boundary condition verification can integrate real-time equipment monitoring data and dynamically adjust safety thresholds to adapt to the performance degradation of aging equipment.

[0089] Compared with related technologies, traditional methods typically use fixed priority rules to allocate heating power, which cannot dynamically respond to changes in equipment health status. In contrast, this solution dynamically adjusts power constraints based on equipment health status, which can extend the service life of critical components.

[0090] Through the above technical solution, this application achieves a dynamic balance between temperature control accuracy, energy efficiency, and equipment operational reliability in the petroleum heating process. It effectively suppresses temperature fluctuations caused by multi-temperature zone thermal interference and avoids equipment overload failures due to improper power distribution. The control command generation process incorporates equipment health status constraints and physical safety limitations, ensuring stable operation of the heating system under optimal conditions while extending the maintenance cycle of key components.

[0091] This application further proposes the following technical solutions, see [link / reference] Figure 5 It includes the following steps:

[0092] 501. Adjust the output power of the electric heating elements in each heating zone according to the temperature control command to control the temperature of each zone to reach the target temperature value;

[0093] 502. Adjust the power distribution strategy of the electric heater according to the energy consumption optimization instructions to shut down or reduce the heating power of non-critical temperature zones;

[0094] 503. Coordinate the working sequence of multiple heaters according to the heater coordination command to avoid the impact on the power supply caused by the simultaneous start-up and shutdown of multiple heaters.

[0095] Among them, the temperature control command refers to the control signal that dynamically adjusts the output power of the electric heating elements in each temperature zone through a closed-loop feedback mechanism. Specifically, a PID controller combined with real-time temperature deviation can be used to determine the power adjustment amount, which is used to eliminate thermal interference between temperature zones and maintain the target temperature. The energy consumption optimization command refers to the dynamic power allocation strategy generated based on the critical temperature zone identification algorithm. Specifically, a priority ranking model can be used to determine non-critical temperature zones, and energy saving can be achieved by reducing or shutting down their power. The heater coordination command refers to the coordination signal that controls the start-up and shutdown sequence of multiple heaters. Specifically, a time-series scheduling algorithm can be used to stagger the start-up and shutdown times of the equipment to avoid the impact of instantaneous current superposition on the power supply.

[0096] Specifically, during the multi-temperature zone heating process of oil flow, the output power of the electric heating elements in each temperature zone is first independently adjusted based on the deviation between the real-time temperature distribution and the target value via temperature control commands. For example, when the temperature in a certain temperature zone is lower than the set value, the power output of the corresponding heating element is increased. Subsequently, non-critical temperature zones, such as sections in a stable flow state, are identified through energy consumption optimization commands, and their heating power is dynamically reduced or completely shut down, thereby reducing energy waste. Finally, the start-up and shutdown sequence of multiple devices is arranged through heater coordination commands. For example, the start-up time interval of heaters in adjacent temperature zones is set as the power recovery cycle to avoid instantaneous power overload caused by the simultaneous start-up of multiple devices.

[0097] Compared to related technologies, traditional methods employ a fixed power allocation model and lack coordinated equipment control, resulting in continuous energy consumption in non-critical temperature zones and power surges caused by the simultaneous start-up and shutdown of multiple heaters. This solution achieves on-demand heating through dynamic power adjustment, combined with a peak-shaving start-up and shutdown strategy to reduce power load fluctuations, optimizing energy consumption and equipment reliability while ensuring temperature control accuracy.

[0098] All of the above-mentioned optional technical solutions can be combined in any way to form the optional embodiments of this application, and will not be described in detail here.

[0099] Through the above technical solutions, this application has solved the problem of insufficient independent temperature control accuracy in multiple temperature zones to a certain extent, and achieved temperature stability in each temperature zone through hierarchical closed-loop control; reduced ineffective energy consumption in non-critical temperature zones, and reduced energy waste through dynamic power allocation strategies; avoided power surges caused by simultaneous start-up and shutdown of multiple devices, and improved power operation stability through timing coordination.

[0100] Figure 6This is a schematic diagram of the structure of an electric heater provided in an embodiment of this application. The electric heater 600 can vary considerably depending on its configuration or performance. It may include one or more Central Processing Units (CPUs) 601 and one or more memories 602. The one or more memories 602 store at least one computer program, which is loaded and executed by the one or more processors 601 to implement the methods provided in the various method embodiments described above. Of course, the electric heater 600 may also have wired or wireless network interfaces, a keyboard, and input / output interfaces for input and output. The electric heater 600 may also include other components for implementing the heater function, which will not be elaborated upon here.

[0101] In an exemplary embodiment, a computer-readable storage medium is also provided, such as a memory including a computer program that can be executed by a processor to perform the heating method in the above embodiments. For example, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage heater, etc.

[0102] In an exemplary embodiment, a computer program product or computer program is also provided, which includes program code stored in a computer-readable storage medium. The processor of the electric heater reads the program code from the computer-readable storage medium and executes the program code, causing the electric heater to perform the heating method described above.

[0103] In some embodiments, the computer program involved in the present application embodiments may be deployed and executed on an electric heater, or on multiple electric heaters located in one location, or on multiple electric heaters distributed in multiple locations and interconnected by a communication network. Multiple electric heaters distributed in multiple locations and interconnected by a communication network may constitute a blockchain system.

[0104] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0105] The above are merely optional embodiments of this application and are not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A heating method, characterized by, The method includes: The system acquires oil heating monitoring data, operating data of electric heaters with independent temperature control in multiple temperature zones, environmental data, and operating condition data. The oil heating monitoring data includes real-time temperature distribution data of multiple heating zones collected by distributed temperature sensors of the electric heater, oil fluid characteristic data collected by the fluid characteristic detection module, and system energy consumption data collected by the energy consumption metering unit. The environmental data includes ambient temperature and ambient humidity, and the operating condition data includes pipeline pressure data and real-time oil flow data. Based on the oil heating monitoring data, the environmental data, the electric heater operation data, and the heater historical maintenance data, the characteristic parameters of the oil heating situation and the characterization parameters of the electric heater operation status are determined. Based on the heating condition characteristic parameters, the operating data, and the heater operating status characterization parameters, temperature control commands, energy consumption optimization commands, and heater coordination commands are determined, including: Based on the heating condition characteristic parameters and the real-time oil flow data in the operating condition data, the oil rheological properties parameters are determined by the oil fluid thermodynamic analysis model. The oil rheological properties parameters include apparent viscosity, flow index and stability coefficient. Based on the heater operating status characterization parameters and petroleum rheological characteristic parameters, a multi-objective collaborative optimization algorithm is used to generate temperature control commands, energy consumption optimization commands, and heater collaborative commands, including: Based on the equipment health status assessment data in the heater operating status characterization parameters and the stability coefficient in the petroleum rheological characteristic parameters, the temperature control priority and power allocation constraints for each heating zone are determined. A constrained reinforcement learning algorithm is used to simultaneously optimize temperature control stability, energy efficiency, and equipment operational reliability, generating preliminary control commands. These preliminary control commands include the target temperature for each temperature zone, the power allocation ratio, and the equipment coordination timing. Conflict detection and boundary condition verification are performed on the preliminary control commands. Conflict detection includes verifying whether the thermal interference between the temperature setpoints of each temperature zone exceeds the allowable range. Boundary condition verification includes checking whether the power allocation exceeds the heater's maximum load capacity and whether the temperature rise rate is within the equipment safety limits. After the preliminary control commands pass the conflict detection and boundary condition verification, they are output as the temperature control command, the energy consumption optimization command, and the heater coordination command. The temperature control command, energy consumption optimization command, and heater coordination command are executed to perform independent temperature control operations on multiple heating sections through which the oil flows.

2. The method of claim 1, wherein, Based on the oil heating monitoring data, the environmental data, the electric heater operation data, and the heater's historical maintenance data, the characteristic parameters of the oil heating situation and the characterization parameters of the electric heater's operating status are determined, including: Based on the oil heating monitoring data and the environmental data, the heating situation characteristic parameters are determined by a multimodal data fusion algorithm. The multimodal data fusion algorithm adopts an adaptive weighted fusion mechanism, which dynamically adjusts the fusion weights according to the data quality of each sensor. Based on the electric heater's operating data and historical maintenance data, the heater's operating status characterization parameters are determined using a heater status assessment model.

3. The method according to claim 2, characterized in that, Based on the oil heating monitoring data and the environmental data, the heating situation characteristic parameters are determined using a multimodal data fusion algorithm, including: The real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data in the oil heating monitoring data are spatiotemporally aligned to generate a spatiotemporally synchronized multi-source data sequence. Based on the multi-source data sequence, the confidence weight of each data source is dynamically determined through the adaptive weighted fusion mechanism. The weight of the real-time temperature distribution data is determined based on its measurement accuracy and real-time performance. The weight of the petroleum fluid characteristic data is determined based on the detection cycle and reliability. The weight of the environmental data is determined based on its correlation with the heating process. A weighted fusion algorithm is used to extract and fuse features from the weighted multi-source data sequence to obtain the heating state feature parameters, which include temperature field distribution features, heat flux density features, and energy efficiency features.

4. The method according to claim 3, characterized in that, The real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data in the petroleum heating monitoring data are spatiotemporally aligned to generate a spatiotemporally synchronized multi-source data sequence, including: The real-time temperature distribution data is preprocessed to extract the temperature measurement values ​​of each heating section at the same time stamp, and the corresponding spatial location information is labeled. The petroleum fluid characteristic data is subjected to time interpolation to align the sampling period with the acquisition time series of the real-time temperature distribution data. The system energy consumption data and environmental data are time-synchronized to ensure that the timestamps are consistent with the timestamps of the real-time temperature distribution data. The real-time temperature distribution data, petroleum fluid characteristic data, system energy consumption data, and environmental data that have been synchronized in time and labeled in space are combined in time series to obtain the spatiotemporally synchronized multi-source data sequence.

5. The method according to claim 2, characterized in that, The electric heater operating data includes heater operating parameters, energy consumption data, and system status data. The heater historical maintenance data includes maintenance records, fault history, and component replacement records. Based on the electric heater operating data and the heater historical maintenance data, the heater operating status characterization parameters are determined through a heater status assessment model, including: A correlation analysis was performed on the heater operating parameters, energy consumption data, and system status data with the maintenance records, fault history, and component replacement records to obtain the correlation analysis results. Based on the correlation analysis results and target matching degree, the heater operating status characterization parameters are determined through the heater status assessment model. The target matching degree is the matching degree between the heater operating parameters and the fault modes in the fault history. The heater operating status characterization parameters include heater operating efficiency indicators, heater energy consumption characteristic parameters, and heater health status assessment data. The heater operating efficiency indicators include heating efficiency coefficient and power conversion efficiency. The heater energy consumption characteristic parameters include energy consumption rate per unit time and energy consumption trend prediction value. The heater health status assessment data includes fault warning level and maintenance urgency indicators.

6. The method according to claim 5, characterized in that, A correlation analysis was performed on the heater operating parameters, energy consumption data, and system status data with the maintenance records, fault history, and component replacement records to obtain the correlation analysis results, including: Extract the load rate, operating temperature and running time time series data from the heater operating parameters, the instantaneous power and cumulative energy consumption time series data from the energy consumption data, and the heater start-stop frequency and operating stability index from the system status data. The runtime time series data, the cumulative energy consumption time series data, and the system status data are dynamically matched with the maintenance cycle and maintenance items in the maintenance record, the operating parameter characteristics and fault type before and after the fault in the fault history, and the component life and performance change data before and after replacement in the component replacement record to obtain the corresponding operating mode similarity and correlation coefficient. The system status data includes heater start-stop frequency and operating stability index. Based on the similarity and correlation coefficient of the operating modes, the correlation analysis results are generated. The correlation analysis results are used to describe the correlation between the current operating status of the electric heater and historical maintenance, faults and component replacement behaviors.

7. The method according to claim 1, characterized in that, Based on the heating condition characteristic parameters and the real-time oil flow data in the operating condition data, the oil rheological characteristic parameters are determined through an oil fluid thermodynamic analysis model, including: The temperature field distribution characteristics, heat flux density characteristics, and real-time oil flow data in the heating state characteristic parameters are input into the oil fluid thermodynamic analysis model. The oil fluid thermodynamic analysis model is a simulation model built based on the principle of non-Newtonian fluid mechanics, including a viscosity-temperature characteristic determination module and a rheological behavior prediction module. The viscosity-temperature characteristic determination module determines the apparent viscosity variation curve of petroleum in different temperature zones based on the temperature field distribution characteristics and petroleum composition data. The rheological behavior prediction module determines the petroleum rheological characteristic parameters based on the apparent viscosity change curve, the heat flux density characteristics, and the real-time petroleum flow data. The petroleum rheological characteristic parameters include apparent viscosity, flow index, and stability coefficient. The petroleum rheological properties parameters are used to characterize the flow properties and stability of petroleum under current heating conditions.

8. An electric heater, characterized in that, The electric heater includes one or more processors and one or more memories, wherein the one or more memories store at least one computer program, which is loaded and executed by the one or more processors to implement the heating method according to any one of claims 1-7.

Images