Accurate temperature control and flow state optimization method for replacing dimethyl disulfide with sulfur in coal tar hydrogenation process

Abstract

The invention provides a temperature accurate control and flow state optimization method for replacing dimethyl disulfide with sulfur in a coal tar hydrogenation process, which comprises the following steps: after acquiring a blockage risk distribution diagram, deploying a flow sensor and a pressure sensor at key nodes of a pipeline in combination with real-time monitoring capability, and collecting dynamic data of a sulfur flow state; flow stability parameters are obtained; according to the fluctuation amplitude, adjusted sulfur flow state data are obtained, the flow velocity and viscosity change of sulfur in the pipeline are calculated through fluid mechanics simulation, and the flow state optimization degree is determined; after a process suitability evaluation result is obtained, the operation cost and the economic benefit improvement range of the sulfur substitution scheme are calculated in combination with a cost accounting model, and the cost control effect is judged by comparing the reference cost of dimethyl disulfide; through a cost control effect, a time sequence prediction algorithm is adopted to analyze sustainability of temperature control and flow state optimization in long-term operation, and a comprehensive optimization scheme of production process stability and economy is obtained.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a method for accurately controlling temperature and optimizing flow state when sulfur replaces dimethyl disulfide in a coal tar hydrogenation process. Background Art

[0002] Background: Coal tar hydrogenation, as a key technology in the coal chemical industry, is of irreplaceable importance in promoting energy conversion and the development of the chemical industry. This process improves the quality and utilization rate of coal tar through hydrogenation, making it a key approach to achieving efficient resource utilization and increased economic benefits. However, process cost and stability have long plagued the industry's development, becoming a key bottleneck restricting further optimization. In traditional processes, dimethyl disulfide, a commonly used injection agent, can meet reaction requirements, but its high market price has led to continuously rising production costs and severely squeezed companies' profit margins. Meanwhile, while the search for low-cost alternatives has become an industry consensus, their practical application has been slow due to technical limitations, making it difficult to achieve the expected results. Existing methods often face insufficient process adaptability when replacing dimethyl disulfide. For example, despite its low price and readily available raw materials, traditional equipment and control methods are unable to effectively address the challenges posed by its physical state changes in actual applications. The molten and liquid sulfur processes are typically separated into separate units, resulting in significant heat loss, complex processes, frequent temperature fluctuations, and a high risk of sulfur solidification, which can lead to pipeline blockages and disrupt production continuity. While some improvements have attempted to optimize the process through integrated units, fundamental breakthroughs in precise temperature control and blockage prevention remain elusive. The core challenges in this area lie in temperature control and flow stability during sulfur applications. Specifically, the molten sulfur zone must maintain a narrow temperature range of 112-120°C, while the liquid sulfur zone must maintain a temperature range of 140-160°C. Any deviation can cause sulfur solidification and equipment failure. Furthermore, inadequate real-time monitoring and control of sulfur flow conditions makes it difficult to effectively predict and mitigate blockage risks. These unresolved technical issues directly lead to low production efficiency and increased cost control difficulties, creating unique challenges for process optimization. Therefore, achieving precise temperature control and optimized flow conditions for sulfur to replace dimethyl disulfide in coal tar hydrogenation processes, while ensuring process stability and economic efficiency, has become a key issue in promoting the efficient development of this technology. Summary of the Invention

[0003] The present invention provides a method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process, which mainly includes: Acquire real-time temperature data from the sulfur melting and liquefaction zones in the process system. Use a sensor array to collect multiple temperature values ​​within the 112-120°C and 140-160°C ranges. Combined with time series analysis, determine the frequency and trend of temperature fluctuations and obtain temperature distribution characteristics. Based on the temperature distribution characteristics, a preset threshold is used to determine the risk of sulfur physical state change. If the temperature is below 112°C or above 160°C, the solidification probability and heat loss degree are calculated through a thermodynamic model to obtain the physical state change prediction result; Based on the physical state change prediction results, pipeline blockage records related to temperature fluctuation frequency are extracted from historical operating data. The conditions and locations of blockages are analyzed using machine learning algorithms to determine the blockage risk distribution map. After obtaining the blockage risk distribution map, combined with real-time monitoring capabilities, flow sensors and pressure sensors are deployed at key nodes in the pipeline to collect dynamic data on the sulfur flow state and obtain flow stability parameters; Based on the flow stability parameters, the control algorithm is used to adjust the heating device power and cooling medium flow rate to maintain the stable range of 112-120°C in the molten sulfur zone and 140-160°C in the liquid sulfur zone, and to determine the fluctuation range after the temperature control is optimized; Based on the fluctuation amplitude, the adjusted sulfur flow state data is obtained, and the flow velocity and viscosity changes of sulfur in the pipeline are calculated through fluid dynamics simulation to determine the degree of flow state optimization; Based on the degree of flow state optimization, real-time indicators of hydrogenation reaction efficiency and production process stability were extracted from the process system. The performance changes after sulfur replaced dimethyl disulfide were analyzed through a data fusion algorithm to obtain process suitability assessment results. After obtaining the process suitability assessment results, the operating costs and economic benefit improvements of the sulfur replacement solution are calculated using a cost accounting model. The cost control effect is determined by comparing the cost with the baseline cost of dimethyl disulfide. Through cost control effects, a time series prediction algorithm is used to analyze the sustainability of temperature control and flow state optimization in long-term operation, and a comprehensive optimization plan for production process stability and economy is obtained.

[0004] The technical solution provided by the embodiment of the present invention may have the following beneficial effects: The present invention discloses a method for temperature control and flow optimization when sulfur replaces dimethyl disulfide in a coal tar hydrogenation process. This method deploys sensor arrays in the sulfur melting and liquefaction zones to collect real-time multi-point temperature data within the ranges of 112-120°C and 140-160°C, and combines this with time series analysis to determine temperature distribution characteristics. Based on the temperature characteristics, the present invention uses preset thresholds to determine the risk of physical state change and calculates the probability of solidification using a thermodynamic model. Based on the physical state change prediction results, the present invention extracts pipeline blockage records from historical data, analyzes the blockage conditions using a machine learning algorithm, and determines the risk distribution. Flow and pressure sensors are deployed at key nodes to collect sulfur flow state data and obtain stability parameters. The present invention uses a control algorithm to adjust the heating device and cooling medium to maintain a stable temperature range, and optimizes the flow state through fluid dynamics simulation. Finally, the present invention integrates indicators such as hydrogenation reaction efficiency to evaluate process suitability and analyzes economic benefits using a cost model. This achieves precise temperature control and flow optimization when sulfur replaces dimethyl disulfide, ensuring the stability and economic efficiency of the coal tar hydrogenation process and providing important technical support for promoting the development of the coal chemical industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0005] Figure 1 The present invention provides a flow chart of a method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process.

[0006] Figure 2 Schematic diagram of a method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process of the present invention.

[0007] Figure 3 This is another schematic diagram of a method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process of the present invention. DETAILED DESCRIPTION

[0008] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0009] like Figure 1-3 In this embodiment, a method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process may specifically include: S101. Acquire real-time temperature data of the sulfur melting zone and liquefaction zone in the process system. Collect multi-point temperature values ​​in the ranges of 112-120°C and 140-160°C through a sensor array. Determine the temperature fluctuation frequency and change trend in combination with time series analysis to obtain temperature distribution characteristics.

[0010] Real-time temperature data is acquired from the sulfur melting and liquefaction zones using a sensor array. Multi-point acquisition is used to cover a specific temperature range, generating a raw temperature dataset. A time series model is constructed based on this raw temperature dataset, analyzing the frequency of temperature fluctuations and identifying periodic variation characteristics. Based on these periodic variation characteristics, a time series decomposition method is used to extract trend components, revealing the long-term temperature variation patterns over time. If these trend components exceed a preset threshold, an anomaly detection mechanism is triggered, comparing historical data to determine whether a process anomaly exists. The anomaly detection results are then analyzed, combined with temperature distribution characteristics, to identify the specific locations of uneven distribution. Based on these spatial temperature differences and the time series trend, regression analysis is used to predict the temperature fluctuation range over a period of time, generating a predicted temperature dataset. If the deviation between the predicted temperature dataset and the real-time temperature data exceeds a preset threshold, adjustment recommendations are generated to determine the optimization direction for the process system operating parameters.

[0011] For example, in temperature monitoring of sulfur melting and liquefaction zones, a sensor array collects real-time temperature data from multiple points, covering the critical range from 100°C to 150°C, to construct a raw data set. Assuming 10 sensors in a given area record the temperature once a minute, a 24-hour data sequence is generated, providing a preliminary understanding of temperature changes over time. This multi-point data collection method captures local variations, ensures comprehensive data, and lays the foundation for subsequent analysis.

[0012] In one possible implementation, when building a time series model based on the original dataset, an autoregressive model can be used to analyze the frequency of temperature fluctuations. Suppose analysis reveals that the temperature exhibits periodic fluctuations every six hours, which may be related to the equipment's operating cycle or environmental factors. After identifying this periodicity, a time series decomposition method is used to break the data into trend, period, and residual components.

[0013] For example, the extracted trend component shows that the temperature slowly rose from 120°C to 130°C over 48 hours, indicating a long-term warming pattern. This decomposition helps focus on long-term changes, ignoring short-term noise and improving analytical accuracy.

[0014] Specifically, if a trend component exceeds a preset threshold, such as the upper limit of 130°C, the anomaly detection mechanism is triggered. By comparing historical data from the past week, if the current heating rate is found to be 30% higher than the historical average, a preliminary judgment can be made that this is a process anomaly. Combined with the anomaly detection results, the spatial temperature differences between the sulfur melting and liquefaction zones are analyzed. For example, the temperature at the center of the melting zone is 135°C, while the edge is only 115°C, a difference of 20°C, indicating uneven distribution in the center. This spatial analysis helps pinpoint the problem location and provides a basis for subsequent adjustments.

[0015] In one possible implementation, regression analysis is used to predict the temperature fluctuation range over the next 12 hours, combining time series trends and spatial differences. Assuming the forecast indicates that the temperature may rise to 140°C, while the real-time data is 138°C, and the deviation does not exceed the preset threshold of 5°C, no adjustment is required. However, if the deviation exceeds the standard, such as the difference between the predicted value and the real-time value reaches 8°C, adjustment suggestion data will be generated, recommending reducing the heating power or increasing the cooling cycle frequency. This mechanism of comparing predictions with real-time can provide early warning of potential risks and ensure process stability. Through the above method, a closed-loop monitoring system is formed from data collection to anomaly detection, and then to prediction and optimization. This not only improves the safety of the sulfur processing process, but also optimizes operating parameters, reduces energy consumption, and brings significant economic benefits and technical advantages.

[0016] S102. Based on the temperature distribution characteristics, a preset threshold is used to determine the risk of sulfur physical state change. If the temperature is lower than 112°C or higher than 160°C, the solidification probability and heat loss degree are calculated using a thermodynamic model to obtain a physical state change prediction result.

[0017] Temperature distribution data is collected in real time through a sensor network. This data is cleaned and formatted to generate a standardized temperature dataset. Based on this standardized temperature dataset, a feature analysis method is used to extract key temperature eigenvalues. Combined with preset thresholds, a preliminary risk assessment is performed to determine whether a physical state change risk exists. If the preliminary risk assessment indicates a physical state change risk, the key temperature eigenvalues ​​are calculated using a thermodynamic model, analyzing the solidification probability and heat loss level to generate a physical state change prediction. Based on this physical state change prediction, combined with the temperature distribution data, a time series analysis method is used to predict the temperature distribution trend over a period of time to determine the potential risk evolution path. Based on this potential risk evolution path and combined with safety range standards, the temperature distribution data is dynamically compared. If anomalies outside the safety range are detected, a risk warning mechanism is triggered to determine the distribution of high-risk areas. The high-risk area distribution data is obtained and, combined with historical temperature distribution and change trends, a support vector machine algorithm is used to classify the risk level, determine the risk severity, and prioritize the response.

[0018] For example, in a sulfur melting and liquefaction process system, when collecting real-time temperature distribution data through a sensor network, a grid layout consisting of multiple sensor nodes can be envisioned, covering the entire process area. Each node collects a temperature value once a minute. Assuming that the temperature readings in a certain area range from 110°C to 150°C, the data cleaning process removes obvious outliers, such as readings below 50°C or above 200°C, and formats the data into a unified timestamp and temperature value pair, forming a standardized temperature dataset. This cleaning and formatting approach enhances the accuracy of subsequent analysis.

[0019] For example, when performing feature analysis on a standardized temperature dataset, key temperature characteristics can be extracted, such as the average temperature, maximum temperature, and temperature fluctuation range within a specific time period. For example, assuming the average temperature of a melting zone is 118°C with a fluctuation range of 5°C, combined with the preset threshold of 120°C and the fluctuation range of 6°C, a preliminary assessment can be made as to whether there is a risk of physical state change. If the average temperature approaches the upper threshold, this may indicate a potential risk, providing a basis for subsequent analysis.

[0020] For example, when analyzing solidification probability and heat dissipation using thermodynamic models, calculations can be made based on key temperature characteristic values ​​and combined with environmental factors such as ambient temperature and equipment heat dissipation conditions. Assuming an ambient temperature of 10°C and rapid heat dissipation from the equipment surface, the model might predict a higher solidification probability for sulfur at the edge of the melting zone. This analysis helps to proactively identify potential physical state changes during the process.

[0021] For example, when using time series analysis to predict temperature distribution trends, based on the temperature data from the past 24 hours, it can be inferred that the temperature in a certain area may drop below 112°C, approaching the freezing point, within the next six hours. This prediction can provide a reference for process adjustments, clarify the evolution of potential risks, and, especially in the event of a continued temperature drop, enable timely countermeasures.

[0022] For example, when dynamically comparing temperature distribution data with safety range standards, assuming a safety range of 115-145°C, if the temperature in a liquefied area drops to 113°C, a risk warning mechanism is triggered. By locating sensor nodes, the high-risk area is determined to be in the northwest corner of the liquefied area. This dynamic comparison and warning mechanism can quickly identify problem areas and improve response efficiency.

[0023] For example, when using a support vector machine algorithm to classify risk levels, historical temperature distribution data and current trends can be combined to categorize risk into high, medium, and low levels. For example, if the temperature in a certain area consistently falls below the safe range and shows a clear downward trend, the algorithm might classify it as high risk and prioritize it. This classification approach helps rationally allocate resources, address temperature anomalies in the process, and ensure stable system operation.

[0024] S103. Based on the prediction results of physical state changes, pipeline blockage records related to temperature fluctuation frequency are extracted from historical operation data. The conditions and locations of blockage occurrence are analyzed through machine learning algorithms to determine the blockage risk distribution map.

[0025] A data set related to temperature fluctuations is obtained from historical operation records. This data set contains information on fluctuation frequency and pipeline blockage events. This data set is filtered using data extraction techniques to obtain a preliminary filtered data set. Based on this preliminary filtered data set, features are extracted for temperature fluctuations and fluctuation frequency, a feature matrix related to pipeline blockages is constructed, and key variables within this feature matrix are identified. A random forest algorithm is used to train this feature matrix, analyzing the correlation between blockage conditions and blockage locations to identify high-risk blockage condition combinations. If these high-risk blockage condition combinations match historical operation records, the matching data is location-tagged to obtain the distribution of high-risk areas associated with the blockage locations. Based on this high-risk area distribution and combined with the changing trends in temperature fluctuation frequency, a risk distribution prediction model is constructed to generate a potential risk distribution map for pipeline blockages. Based on this potential risk distribution map, data is verified for high-risk areas. If the deviation between the verification results and historical blockage locations exceeds a preset threshold, the weights in the feature matrix are adjusted to update the risk distribution map. This updated risk distribution map is then dynamically updated with the latest temperature fluctuation data to determine the priority monitoring areas for current pipeline blockages.

[0026] For example, when analyzing the relationship between temperature fluctuations and pipeline blockages, relevant data can be extracted from historical operating records, focusing on the potential connection between the frequency of temperature fluctuations and blockage events. To filter data sets, a fluctuation frequency threshold can be set, such as data with more than five fluctuations per hour, as a preliminary screening criterion to eliminate irrelevant noise data and ensure the accuracy of subsequent analysis. This approach helps focus on high-risk scenarios and avoids interference caused by data redundancy.

[0027] For example, during the feature extraction phase, a feature matrix can be constructed to analyze temperature fluctuations and their frequency, extracting key variables such as fluctuation amplitude, duration, and period. For example, if the temperature in a certain area of ​​a pipeline fluctuates by 10 degrees Celsius within 24 hours, with a period of once every two hours, this high frequency and amplitude feature is likely highly correlated with a blockage event. By incorporating these variables into the feature matrix, a clear data foundation can be provided for subsequent analysis.

[0028] For example, when training the feature matrix using the random forest algorithm, historical data can be used to analyze the correlation between blockage conditions and locations. For example, if a pipeline experiences temperature fluctuations six times per hour and the amplitude exceeds 8 degrees Celsius, the probability of blockage increases significantly. This combination of conditions can be labeled as high-risk. By matching historical records, it can be found that a certain section of pipeline has been blocked repeatedly under similar conditions, thus marking it as a high-risk area. This method helps to accurately locate problem areas.

[0029] For example, when building a risk distribution prediction model, we can combine trends in temperature fluctuation frequency to create a potential risk distribution map. For example, if the fluctuation frequency in a pipeline area increased from 3 to 7 times per hour over the past week, combined with historical congestion data, we can predict a high congestion risk in that area within the next 24 hours. This predictive map provides a visual basis for monitoring key areas.

[0030] For example, during data verification, if a predicted high-risk area deviates significantly from the historical congestion location—for example, if the predicted area deviates by more than 50 meters from the actual congestion point—then the weights for the frequency or amplitude of fluctuations in the feature matrix are adjusted and the risk distribution map is regenerated. This dynamic adjustment mechanism continuously optimizes prediction results and improves reliability.

[0031] For example, when dynamically updating risk distribution prediction results, the system can combine the latest temperature fluctuation data. If the current fluctuation frequency in a certain area suddenly increases to 8 times per hour, the system can immediately update the risk distribution and determine that area as a priority monitoring target. This real-time performance helps to promptly identify potential problems and reduce pipeline operation risks.

[0032] For example, after determining priority monitoring areas, the density of temperature sensors can be increased in high-risk areas, such as reducing the sensor spacing from 100 meters to 50 meters to obtain more detailed fluctuation data. This measure can further improve monitoring accuracy and support subsequent risk prevention and control.

[0033] S104. After obtaining the blockage risk distribution map, flow sensors and pressure sensors are deployed at key nodes of the pipeline in combination with real-time monitoring capabilities to collect dynamic data on the sulfur flow state and obtain flow stability parameters.

[0034] By deploying flow sensors and pressure sensors at key nodes of the pipeline, dynamic data of sulfur flow is obtained to obtain an original state data set. Based on the original state data set, pre-processing technology is used to denoise and standardize the data to obtain a clean data set. If the flow data or pressure data in the clean data set exceeds the preset threshold range, an abnormal mark is triggered, and the abnormal time period and the corresponding node location information are determined. Based on the abnormal time period and the node location information, combined with the pre-established congestion risk distribution map, the support vector machine algorithm is used to classify the abnormal data to determine the potential congestion risk level. Based on the congestion risk level obtained by the classification, the changing trend of the historical flow stability parameters of the high-risk nodes is obtained to determine the dynamic characteristics of the risk evolution. If the dynamic characteristics show that the risk evolution trend continues to deteriorate, the current flow stability parameters are calculated in combination with the real-time monitoring data to obtain the latest stability assessment results. Based on the latest stability assessment results, the monitoring frequency and sensor data acquisition density are adjusted to generate an optimized data acquisition strategy for the high-risk nodes.

[0035] For example, during the flow and pressure data collection process at key pipeline nodes, dynamic information on sulfur flow can be obtained by deploying high-precision sensors. Suppose that at a key node in a certain pipeline section, the flow sensor's measurement range is set to 0.5 to 5.0 cubic meters per minute, and the pressure sensor's range is 0.2 to 2.0 MPa. When the data exceeds these preset thresholds, the system automatically flags the anomaly. For example, between 10:00 and 11:00 a.m. on a certain day, the flow rate at a certain node plummets to 0.3 cubic meters per minute, while the pressure rises to 2.5 MPa, clearly exceeding the threshold range. The system immediately records this time period and node location, triggering an anomaly flag. This approach can quickly locate potential problem areas and provide a precise basis for subsequent analysis.

[0036] For example, during the data preprocessing stage, denoising and standardization are key steps in ensuring data quality. Denoising can filter out invalid data caused by sensor jitter or environmental interference. For example, during sulfur pipeline operation, equipment vibration may cause brief fluctuations in flow data. Smoothing can eliminate these noises. Standardization unifies data of different dimensions to the same scale for ease of subsequent analysis. For example, assuming the raw flow data ranges from 0.5 to 5.0 and the pressure data ranges from 0.2 to 2.0, both are adjusted to the range of 0 to 1 after standardization to facilitate algorithm processing. This processing method helps improve the accuracy of data analysis.

[0037] For example, using a support vector machine algorithm to classify abnormal data can effectively distinguish different risk levels. Based on historical data, if a node's flow rate drops by more than 30% or its pressure rises by more than 20% during an abnormal time period, combined with the congestion risk distribution map, the node may be classified as high-risk. This classification method extracts features from multi-dimensional data to determine risk levels, providing support for subsequent decision-making.

[0038] For example, when analyzing flow stability parameters at high-risk nodes, one can focus on the frequency of flow and pressure fluctuations in historical data. Suppose that over the past month, flow at a certain node fluctuated three times per day and pressure five times per day. Recently, this frequency has increased to over eight times per day, indicating a worsening risk trend. In this case, the current stability parameters are calculated using real-time data. If the fluctuation frequency continues to increase, monitoring frequency should be increased. This dynamic analysis method helps to promptly capture changes in risk.

[0039] For example, when adjusting monitoring strategies, data collection density can be increased for high-risk nodes. For example, suppose the original monitoring frequency is once an hour, but now it is adjusted to once every 15 minutes. Temporary sensors can also be added near these nodes to ensure more comprehensive data coverage. This optimization strategy can improve monitoring precision and provide more reliable assurance for pipeline operations.

[0040] S105. Using the flow stability parameters, a control algorithm is used to adjust the power of the heating device and the flow rate of the cooling medium, to maintain stable ranges of 112-120°C and 140-160°C in the molten sulfur zone and the liquid sulfur zone, respectively, and to determine the fluctuation amplitude after the temperature control is optimized.

[0041] Sensors collect real-time flow stability data from the molten sulfur and liquid sulfur zones, acquiring dynamic temperature information within these zones and determining the initial temperature distribution. Based on this initial temperature distribution, a preset control algorithm is used to calculate power adjustment values ​​for the heating device and flow adjustment values ​​for the cooling medium, generating control parameters for the two zones. These control parameters are transmitted to the heating device and cooling medium adjustment modules, which perform power and flow adjustment operations and obtain real-time adjusted temperature data. If the real-time adjusted temperature data exceeds a preset temperature range, the control algorithm recalculates the adjustment parameters and transmits them to the relevant modules for secondary adjustment to determine whether the temperature has returned to a stable range. If the temperature has returned to a stable range, the flow stability data is continuously monitored, and the dynamic fluctuation range is determined by combining historical temperature trends to determine the effectiveness of fluctuation control. Based on this fluctuation control effectiveness data, the stable control state of the molten sulfur and liquid sulfur zones is analyzed, and long-term temperature fluctuation predictions are obtained using time series analysis. Based on these predictions, if the long-term fluctuation range exceeds a preset threshold, the control algorithm parameters are adjusted, the power adjustment values ​​and flow adjustment values ​​are recalculated, and the optimized stable control state is determined.

[0042] For example, when collecting flow stability data for the molten sulfur and liquid sulfur areas, high-precision temperature sensors can be deployed to provide real-time information on temperature changes within both areas. Suppose the initial temperature distribution in the molten sulfur area shows a core temperature of 135°C, while the edge temperature is 120°C. This significant temperature difference could affect the flow stability of the sulfur. To address this situation, a pre-set control algorithm, based on the temperature differential data, calculates that the heating system power needs to be increased to 80% of its rated value, while simultaneously reducing the cooling medium flow rate to 2 liters per minute to balance the overall temperature distribution.

[0043] Specifically, after the control parameters are transmitted to the heating device and cooling medium adjustment module, power and flow adjustments are performed. It can be observed that the core temperature of the molten sulfur area gradually rises to 130 degrees Celsius, and the temperature of the edge area is also adjusted to 125 degrees Celsius, narrowing the temperature difference. If the real-time adjusted temperature data exceeds the preset range, for example, the core area temperature reaches 140 degrees Celsius, exceeding the upper limit by 5 degrees, the control algorithm recalculates the adjustment parameters, reduces the heating power to 60% of the rated value, and increases the cooling medium flow to 3 liters per minute, performing a secondary adjustment until the temperature returns to the stable range of 125 to 130 degrees Celsius.

[0044] For example, after the temperature returned to a stable range, continuous monitoring of flow stability data, combined with historical temperature trends, revealed that the temperature fluctuation range in the molten sulfur area had narrowed from 3°C to 1.5°C over the previous 24 hours, indicating effective fluctuation control. Time series analysis predicted that temperature fluctuations would likely remain within 1.8°C over the next 48 hours, within the preset threshold of 2°C, eliminating the need to adjust control algorithm parameters. This predictive analysis helps identify temperature trends in advance and ensure system stability.

[0045] Specifically, if the forecast indicates long-term fluctuations exceeding a preset threshold—for example, a potential fluctuation of 2.5 degrees Celsius within the next 72 hours—the control algorithm parameters will be adjusted. For example, the algorithm's sensitivity to temperature fluctuations will be increased, the heating power adjustment value will be recalculated to 70% of the rated value, and the cooling medium flow rate will be adjusted to 2.5 liters per minute to optimize stable control. This dynamic adjustment method effectively addresses potential risks of temperature runaway.

[0046] For example, when analyzing the stable control states of the molten sulfur and liquid sulfur areas, it was found that the liquid sulfur area, due to its greater fluidity, typically experiences greater temperature fluctuations than the molten sulfur area, requiring more frequent adjustments to the cooling medium flow rate. By setting different control parameters for the two areas, for example, setting the initial cooling medium flow rate to 3.5 liters per minute in the liquid sulfur area and 2 liters in the molten sulfur area, this strategy better accommodates the characteristics of each area. This differentiated control strategy improves the adaptability and stability of the overall system, providing reliable assurance for sulfur pipeline operations.

[0047] S106. According to the fluctuation amplitude, the adjusted sulfur flow state data is obtained, and the flow velocity and viscosity changes of the sulfur in the pipeline are calculated through fluid mechanics simulation to determine the degree of flow state optimization.

[0048] Initial flow state information is extracted from the fluctuation amplitude data of the sulfur in the pipeline and compared with a pre-established database to obtain preliminary flow characteristics of the sulfur in the pipeline environment. Based on this preliminary flow characteristics, the flow velocity distribution and viscosity parameter changes of the sulfur in the pipeline are obtained. Multi-dimensional calculations are performed using a fluid dynamics simulation tool to determine the dynamic distribution of flow velocity and viscosity. Based on this dynamic distribution, the degree to which the fluctuation amplitude affects the flow velocity distribution is analyzed. If the fluctuation amplitude exceeds a preset threshold, the simulation parameters are adjusted to obtain optimized flow velocity distribution data. Based on this optimized flow velocity distribution data, the interaction between viscosity changes and the pipeline environment is calculated. If the viscosity parameter deviates from a preset range, finite element analysis is used to calibrate it to obtain an adjusted viscosity parameter value. Using this adjusted viscosity parameter value, combined with the fluid dynamics simulation results, the correlation between the flow state and the degree of optimization is analyzed to determine the direction for improving the flow state of the sulfur in the pipeline. Based on this improvement direction, a support vector machine algorithm is used to comprehensively predict the fluctuation amplitude, flow velocity distribution, and viscosity parameters to determine the final optimization degree data.

[0049] For example, when analyzing sulfur fluctuations within a pipeline, sensors can collect real-time flow rate information at different locations within the pipeline to initially extract flow state data. For example, suppose the sulfur fluctuations in a certain section of the pipeline are characterized by flow rate variations of 5 to 10 tons per minute. This fluctuation could be due to pipeline wall roughness or localized pressure changes. Based on this, the collected data is compared with a pre-established database containing flow characteristic models corresponding to different fluctuation amplitudes. This database then provides preliminary insights into the sulfur flow characteristics under the current pipeline conditions, such as the presence of periodic pulsation.

[0050] In one possible implementation, the flow velocity distribution and viscosity parameter changes can be determined by measuring the sulfur flow velocity at different sections of the pipeline using an ultrasonic velocimeter. Assuming a flow velocity of 2.5 meters per second at the pipeline inlet and dropping to 1.8 meters per second in the middle section, the viscosity value measured by viscosity testing equipment increases from 500 centipoise to 600 centipoise, reflecting the increased resistance to sulfur flow. Subsequently, using fluid dynamics simulation tools to simulate the dynamic distribution of flow velocity and viscosity, the authors analyzed the possibility that the decreased flow velocity and increased viscosity may have caused localized vortices in the middle section of the pipeline.

[0051] For example, in analyzing the impact of fluctuation amplitude on velocity distribution, if the fluctuation amplitude exceeds a preset threshold, such as a flow rate change exceeding 8 tons per minute, simulation parameters can be adjusted, such as increasing the assumed value for pipe inner wall smoothness. After recalculation, the optimized velocity distribution data can be obtained, and the inlet velocity may stabilize at 2.3 meters per second. This adjustment helps reduce instability caused by sudden changes in local velocity.

[0052] In one possible implementation, if the measured viscosity parameter deviates from a preset range, such as exceeding 650 centipoise, due to the interaction between viscosity changes and pipeline environment, finite element analysis can be used to calibrate the viscosity to simulate the viscosity behavior of sulfur at different temperatures. This adjustment results in a viscosity value closer to reality, such as 620 centipoise. This calibration more realistically reflects the flow resistance characteristics of sulfur in the pipeline.

[0053] For example, when analyzing the correlation between flow state and optimization level, combining fluid dynamics simulation results with adjusted viscosity parameters, it was found that reduced viscosity resulted in a more uniform flow velocity distribution. This led to the identification of a direction for improvement: optimizing the temperature distribution within the pipe to control viscosity. This analysis provided clear guidance for subsequent process adjustments.

[0054] In one possible implementation, a support vector machine algorithm can be used to comprehensively predict fluctuation amplitude, velocity distribution, and viscosity parameters. Historical data can be input. For example, if the average fluctuation amplitude over the past hour was 7 tons, the average velocity was 2.2 meters per second, and the average viscosity was 610 centipoise, the prediction would be that the fluctuation amplitude might drop to 6 tons within the next hour. This prediction can help adjust pipeline operating parameters in advance, maintain flow stability, reduce equipment wear risks, and improve sulfur transportation efficiency.

[0055] S107. Based on the degree of flow state optimization, real-time indicators of hydrogenation reaction efficiency and production process stability are extracted from the process system. The performance changes after sulfur replaces dimethyl disulfide are analyzed through a data fusion algorithm to obtain the process suitability evaluation results.

[0056] Multi-source data from the hydrogenation reaction process is acquired, including real-time indicators related to flow state, to construct an initial data set. Based on the initial data set, a data fusion algorithm is used to integrate the multi-source data to obtain a comprehensive characteristic value for hydrogenation reaction efficiency and production process stability. Based on the comprehensive characteristic value, the performance change trend after sulfur replaces dimethyl disulfide is analyzed to determine whether the performance change trend exceeds a preset optimization threshold. If the performance change trend exceeds the preset optimization threshold, the corresponding flow state parameter is recorded. Based on the flow state parameter and the changes in the real-time indicators, the fluctuation range of the stability indicator of the production process is analyzed to determine whether the stability indicator meets the process system requirements. If the stability indicator meets the process system requirements, the improvement in hydrogenation reaction efficiency is compared with the results of the data fusion algorithm to obtain a preliminary conclusion on process suitability. Based on the preliminary conclusion, the process suitability assessment results are verified using a support vector machine algorithm, combining the comprehensive analysis of the performance change trend and the stability indicator, to obtain final suitability assessment data.

[0057] For example, when acquiring multi-source data during the hydrogenation reaction, real-time flow-related metrics, such as pipeline pressure, temperature, and flow rate, can be collected from multiple equipment sensors. Assuming a production scenario with a pressure of 2.5 MPa, a temperature of 300°C, and a flow rate of 1.2 m / s, these data constitute the initial dataset. This initial compilation of data lays the foundation for subsequent analysis.

[0058] For example, after constructing the initial dataset, data fusion algorithms can be used to integrate multi-source data. This can standardize the data from different sources to eliminate dimensional differences. For example, pressure and temperature data can be fused using a weighted average method to obtain an eigenvalue that comprehensively reflects the flow state. Assuming the fused eigenvalue is 0.75, this indicates that the current reaction system is operating in a relatively stable state. This method can effectively improve the overall consistency of the data and provide a reliable basis for subsequent analysis.

[0059] For example, when analyzing the performance trends after replacing dimethyl disulfide with sulfur based on comprehensive characteristic values, the evaluation can be conducted from two perspectives: reaction efficiency and product purity. Suppose, after substitution, reaction efficiency increases from 85% to 88%, while product purity decreases slightly. This trend needs to be compared with a preset threshold of 0.8. If the threshold is exceeded, the corresponding flow rate and pressure parameters are recorded for further analysis. This analysis method helps to identify potential problems promptly.

[0060] For example, when determining the fluctuation range of a production process stability indicator, one can focus on whether the pressure fluctuation is within the range of ±0.2 MPa. If the fluctuation range meets the requirements, for example, the actual fluctuation is only ±0.1 MPa, then the system is operating smoothly. This judgment method can provide an important reference for process optimization.

[0061] For example, comparing the efficiency gains of hydrogenation reactions, data fusion revealed a 3 percentage point increase. This comparative analysis helps to initially determine whether the process suitability meets expectations and provides data support for subsequent verification.

[0062] For example, when using a support vector machine algorithm to validate process suitability assessment results, flow state parameters and stability indicators can be used as input features to predict the final suitability score. Assuming a predicted score of 0.9 indicates that the current process adjustment direction is reasonable, this validation method can improve the scientific nature and accuracy of the assessment and help optimize production processes.

[0063] For example, if a comprehensive analysis of performance trends and stability indicators shows positive results, such as increased efficiency and reduced fluctuations, the process adjustment can be considered highly adaptable. This multi-dimensional, comprehensive analysis can provide comprehensive support for production decisions and ensure efficient and stable system operation.

[0064] S108. After obtaining the process suitability assessment results, calculate the operating costs and economic benefit improvements of the sulfur replacement solution using the cost accounting model, and determine the cost control effect by comparing it with the baseline cost of dimethyl disulfide.

[0065] Process suitability assessment data is obtained from a database and de-noised and formatted using pre-established data cleaning rules to obtain a pre-processed assessment data set. Based on this pre-processed assessment data set, a cost accounting model is used to calculate the operating costs of the sulfur substitution solution and determine detailed cost component data. Based on this detailed cost component data and incorporating economic benefit assessment logic, the economic benefit improvement of the sulfur substitution solution is calculated to obtain a corresponding benefit index value. This benefit index value is compared with pre-set benchmark cost data to determine whether the operating cost is lower than the benchmark cost. If so, the cost control is marked as effective. If the cost control is marked as effective, the key factors influencing the economic benefit improvement are further extracted and weighted using a regression analysis model to determine the primary cost drivers. Based on these primary cost drivers, optimized adjustment parameters for the sulfur substitution solution are generated, and the adjusted operating cost and economic benefit forecast values ​​are output.

[0066] For example, when acquiring process suitability assessment data from a database, a specially designed query tool can be used to extract historical records related to hydrogenation reactions. This data may include reaction efficiency and stability indicators under different flow conditions, covering multi-dimensional information before and after sulfur replaces dimethyl disulfide. After acquiring the data, pre-established data cleaning rules, such as removing outliers and standardizing timestamp formats, are applied to ensure data consistency, laying the foundation for subsequent analysis.

[0067] For example, when using a cost accounting model for a cleaned evaluation dataset, operating costs can be broken down into raw material costs, energy costs, and equipment maintenance costs. Assuming raw material costs account for 40% and energy costs account for 35%, model analysis reveals that sulfur substitution offers significant potential for energy savings. This detailed analysis helps clarify the direction of cost control.

[0068] For example, in the application of economic benefit evaluation logic, the economic benefit improvement can be calculated by comparing the average annual revenue of an alternative solution with that of a traditional solution. Assume that the average annual revenue of the traditional solution is 1 million yuan, while the alternative solution increases it to 1.2 million yuan, a 20% improvement. This indicator directly reflects the economic value of the solution and provides a basis for subsequent decision-making.

[0069] For example, when comparing operating costs against benchmark costs, assuming the benchmark cost is 5,000 yuan per ton of product, and the operating cost of the alternative is 4,800 yuan, which is lower than the benchmark, then the alternative is marked as cost-effective. This marking method makes it easy to quickly screen out solutions with economic advantages.

[0070] For example, when further analyzing the key factors influencing the economic benefits of effective cost control solutions, a regression analysis model revealed that raw material price fluctuations had the highest impact on costs, accounting for 50%. This result suggests that optimizing raw material procurement strategies may be the key to reducing costs.

[0071] For example, generating optimized adjustment parameters based on key cost drivers can suggest bulk purchasing or supplier optimization strategies for raw material procurement costs. Assuming the adjusted operating costs drop to 4,700 yuan per ton, the projected annual economic benefits increase to 1.25 million yuan, demonstrating significant optimization results. The output of these adjustment parameters provides concrete guidance for process improvement.

[0072] For example, in the logical progression from the core solution to the extended solution, the core solution can be cost optimization based on existing data, while the extended solution can consider introducing lower-cost alternative raw materials or improving equipment operation modes to further improve economic benefits. Such a multi-level design not only ensures the feasibility of the solution, but also expands the possibility of application scenarios. Through the above multi-faceted analysis and examples, it can be seen that each step from data acquisition to cost optimization is closely centered on the evaluation of the suitability of the hydrogenation reaction process. Each link supports each other and jointly provides support for improving economic benefits and process stability. The application of these methods not only helps with cost control, but also provides a data-driven decision-making basis for the continuous improvement of process systems.

[0073] S109. Through cost control effects, a time series prediction algorithm is used to analyze the sustainability of temperature control and flow state optimization in long-term operation, and a comprehensive optimization plan for production process stability and economy is obtained.

[0074] Temperature control and flow state data from the production process are acquired from a historical data acquisition system and organized into a time series format to generate an initial dataset. Based on this initial dataset, an autoregressive integrated moving average model, part of the time series prediction method, is used to analyze the changing trends of temperature control and flow state, determining the fluctuation range of key influencing factors. If the fluctuation range of these key influencing factors exceeds a preset threshold, a data filtering module cleans out outliers to generate a cleaned dataset, and the stability of these influencing factors is determined. This cleaned dataset, combined with cost control indicator data, analyzes the impact of temperature control and flow state on economic efficiency, generating a cost-effectiveness correlation model. Based on this cost-effectiveness correlation model, an optimization algorithm is used to adjust the parameters of temperature control and flow state, obtaining the adjusted parameter combination and determining the optimal configuration for the production process. If the optimized configuration meets the stability criteria during simulation, the data validation module performs multiple rounds of testing on the optimized configuration to obtain a final stability assessment result. Based on this final stability assessment result, an execution strategy for the comprehensive optimization solution is generated, resulting in a production process adjustment plan suitable for long-term operation.

[0075] For example, in production, data collection on temperature control and flow conditions is fundamental to optimizing production configurations. Suppose that temperature and flow rate data for a chemical production line over the past month has been extracted from a historical data acquisition system, recorded hourly in a time series format. The temperature data ranges from 50 to 80 degrees Celsius, and the flow rate data ranges from 200 to 500 liters per minute. This data reflects the operating status of production equipment under varying loads, laying the foundation for subsequent analysis.

[0076] For example, the autoregressive integrated moving average model, a time series forecasting method, can be understood as a way to predict future trends using historical data. Suppose analysis reveals that temperature shows a periodic upward trend over a specific time period, while flow rate exhibits irregular fluctuations. This prediction suggests that the key influencing factor for temperature fluctuations may be ambient temperature differences, while flow rate fluctuations may be related to the frequency of equipment valve adjustments. If the fluctuation range exceeds a preset threshold, such as a temperature fluctuation exceeding 5 degrees Celsius or a flow fluctuation exceeding 50 liters per minute, data cleaning is required to remove outliers and ensure the accuracy of subsequent analysis.

[0077] For example, after data cleaning, the stability of influencing factors can be assessed by calculating the standard deviation of the cleaned data. Assuming the standard deviation of the cleaned temperature data is 2.3 degrees Celsius and the standard deviation of the flow rate is 15 liters per minute, both within acceptable ranges, indicating high data stability. This stability provides a reliable basis for subsequent cost-effectiveness analysis.

[0078] For example, when analyzing the impact of temperature and flow conditions on economic efficiency in conjunction with cost control indicators, suppose we find that for every 1°C increase in temperature, energy costs increase by 0.5 yuan per hour, while a 10 liters per minute increase in flow rate improves raw material utilization by 0.2%. By building a correlation model, we can clearly see that temperature control plays a dominant role in cost, while flow optimization tends to improve efficiency. This analysis helps find the right balance between cost and effectiveness.

[0079] For example, during the parameter adjustment phase, an optimization algorithm was used to adjust temperature and flow parameters. Assuming the temperature control target was set to 65 degrees Celsius and the flow rate to 350 liters per minute, a simulation revealed that energy costs decreased by approximately 8% and raw material utilization increased by approximately 5%. Once this optimized configuration met stability requirements in the simulation, it required multiple rounds of testing and verification, such as running it continuously for 48 hours to observe any abnormal fluctuations.

[0080] For example, when generating an execution strategy for a comprehensive optimization plan, assume the ultimate plan is to upgrade temperature control equipment to a more precise model and adjust the flow valve response time to less than 5 seconds. This strategy can maintain long-term production stability while effectively reducing operating costs and improving overall economic benefits. Through comprehensive data support and parameter adjustments, the optimization plan has demonstrated strong adaptability and sustainability in practical applications.

[0081] The above disclosure is only a preferred embodiment of the present invention, and it is certainly not intended to limit the scope of the present invention. A person skilled in the art can understand that all or part of the processes of the above embodiment and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process, characterized in that: The method comprises: Acquire real-time temperature data from the sulfur melting and liquefaction zones in the process system. Use a sensor array to collect multiple temperature values ​​within the 112-120°C and 140-160°C ranges. Combined with time series analysis, determine the frequency and trend of temperature fluctuations and obtain temperature distribution characteristics. Based on the temperature distribution characteristics, a preset threshold is used to determine the risk of sulfur physical state change. If the temperature is below 112°C or above 160°C, the solidification probability and heat loss degree are calculated through a thermodynamic model to obtain the physical state change prediction result; Based on the physical state change prediction results, pipeline blockage records related to temperature fluctuation frequency are extracted from historical operating data. The conditions and locations of blockages are analyzed using machine learning algorithms to determine the blockage risk distribution map. After obtaining the blockage risk distribution map, combined with real-time monitoring capabilities, flow sensors and pressure sensors are deployed at key nodes in the pipeline to collect dynamic data on the sulfur flow state and obtain flow stability parameters; Based on the flow stability parameters, the control algorithm is used to adjust the heating device power and cooling medium flow rate to maintain the stable range of 112-120°C in the molten sulfur zone and 140-160°C in the liquid sulfur zone, and to determine the fluctuation range after the temperature control is optimized; Based on the fluctuation amplitude, the adjusted sulfur flow state data is obtained, and the flow velocity and viscosity changes of sulfur in the pipeline are calculated through fluid dynamics simulation to determine the degree of flow state optimization; Based on the degree of flow state optimization, real-time indicators of hydrogenation reaction efficiency and production process stability were extracted from the process system. A data fusion algorithm was used to analyze the performance changes after sulfur replaced dimethyl disulfide, resulting in a process suitability assessment. After obtaining the process suitability assessment results, the operating costs and economic benefit improvements of the sulfur replacement solution were calculated using a cost accounting model. The cost control effect was then determined by comparing the cost with the baseline cost of dimethyl disulfide. Through cost control effects, a time series prediction algorithm is used to analyze the sustainability of temperature control and flow state optimization in long-term operation, and a comprehensive optimization plan for production process stability and economy is obtained.

2. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: The method acquires real-time temperature data of the sulfur melting zone and liquefaction zone in the process system, collects multiple temperature values ​​in the ranges of 112-120°C and 140-160°C through a sensor array, and determines the temperature fluctuation frequency and change trend in combination with time series analysis to obtain temperature distribution characteristics, including: Real-time temperature data is acquired from the sulfur melting and liquefaction zones through a sensor array. Multi-point acquisition is used to cover a specific temperature range to obtain the original temperature data set. Building a time series model based on the original temperature data set, analyzing the fluctuation frequency of the temperature data, and determining the periodic change characteristics; In view of the periodic change characteristics, the time series decomposition method is used to extract the change trend components and obtain the long-term change law of temperature over time; If the change trend component exceeds the preset threshold range, the anomaly detection mechanism is triggered to determine whether there is a process anomaly by comparing historical data; Obtaining the results of the anomaly detection, analyzing the spatial temperature difference between the sulfur melting zone and the liquefaction zone in combination with the temperature distribution characteristics, and determining the specific location of the uneven distribution; Based on the results of the spatial temperature difference and the change trend of the time series, a regression analysis method is used to predict the temperature fluctuation range in the future to obtain a predicted temperature data set; If the deviation between the predicted temperature data set and the real-time temperature data exceeds a preset threshold, adjustment suggestion data is generated to determine the optimization direction of the process system operating parameters.

3. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: Based on the temperature distribution characteristics, a preset threshold is used to judge the risk of sulfur physical state change. If the temperature is lower than 112°C or higher than 160°C, the solidification probability and heat loss degree are calculated through a thermodynamic model to obtain the physical state change prediction results, including: collecting temperature distribution data in real time through a sensor network, and performing cleaning and formatting processing on the temperature distribution data to obtain a standardized temperature data set; Based on the standardized temperature data set, a feature analysis method is used to extract key temperature characteristic values, and a preliminary risk assessment is performed based on the preset threshold to determine whether there is a risk of physical state change; If the preliminary risk assessment result indicates that there is a risk of physical state change, the key temperature characteristic value is calculated using a thermodynamic model, and the solidification probability and heat loss degree are analyzed to obtain a physical state change prediction result; Based on the physical state change prediction results and the temperature distribution data, a time series analysis method is used to predict the temperature distribution trend in the future to obtain the potential risk evolution path; According to the potential risk evolution path and combined with the safety range standard, the temperature distribution data is dynamically compared. If an abnormal point exceeding the safety range is found, the risk warning mechanism is triggered to determine the distribution of high-risk areas; The distribution data of the high-risk areas is obtained, and combined with the historical temperature distribution and change trend, the support vector machine algorithm is used to classify the risk level, determine the severity of the risk and the priority order.

4. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: Based on the physical state change prediction results, pipeline blockage records related to temperature fluctuation frequency are extracted from historical operation data. The conditions and locations of blockage occurrence are analyzed through machine learning algorithms to determine the blockage risk distribution map, including: Acquire a data set related to temperature fluctuations from historical operation records, wherein the data set includes information on fluctuation frequencies and pipeline blockage events; Screening the data set using a data extraction technique to obtain a preliminary screening data set; Based on the preliminary screening data set, feature extraction is performed on temperature fluctuation and fluctuation frequency, a feature matrix related to pipeline blockage is constructed, and key variables in the feature matrix are determined; A random forest algorithm is used to train the feature matrix, analyze the correlation pattern between congestion conditions and congestion locations, and determine high-risk congestion condition combinations; If the high-risk congestion condition combination matches the historical operation record, the matching data is position-marked to obtain the high-risk area distribution related to the congestion location; Based on the distribution of high-risk areas and the trend of temperature fluctuation frequency changes, a risk distribution prediction model is constructed to obtain a potential risk distribution map of pipeline blockage; Based on the potential risk distribution map, data verification is performed for high-risk areas. If the deviation between the verification result and the historical congestion location exceeds a preset threshold, the weights in the feature matrix are adjusted to update the risk distribution map. The updated risk distribution map is obtained, and combined with the latest temperature fluctuation data, the risk distribution prediction result is dynamically refreshed to determine the priority monitoring area of ​​the current pipeline blockage.

5. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: After obtaining the blockage risk distribution map, flow sensors and pressure sensors are deployed at key nodes of the pipeline in combination with real-time monitoring capabilities. By deploying flow sensors and pressure sensors at key nodes of the pipeline, dynamic data of sulfur flow is obtained to obtain the original state data set; According to the original state data set, a preprocessing technology is used to remove noise and standardize the data to obtain a clean data set; If the flow data or pressure data in the clean data set exceeds a preset threshold range, an abnormal flag is triggered, and the abnormal time period and corresponding node location information are determined; Based on the abnormal time period and the node location information, combined with a pre-established congestion risk distribution map, a support vector machine algorithm is used to classify the abnormal data to determine the potential congestion risk level; Based on the congestion risk level obtained by the classification, the change trend of the historical flow stability parameters of the high-risk nodes is obtained to determine the dynamic characteristics of the risk evolution; If the dynamic characteristics show that the risk evolution trend continues to worsen, the current flow stability parameters are calculated in combination with the real-time monitoring data to obtain the latest stability assessment results; According to the latest stability assessment results, the monitoring frequency and sensor data collection density are adjusted, and an optimized data collection strategy is generated for the high-risk nodes.

6. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: The control algorithm is used to adjust the power of the heating device and the flow rate of the cooling medium by using the flow stability parameter to maintain the stable range of 112-120°C and 140-160°C in the molten sulfur zone and the liquid sulfur zone respectively, and to determine the fluctuation range after the temperature control is optimized, including: The flow stability data of the molten sulfur area and the liquid sulfur area are collected in real time by sensors to obtain the dynamic information of temperature changes in the areas and determine the initial temperature distribution state; According to the initial temperature distribution state, a preset control algorithm is used to calculate the power adjustment value of the heating device and the flow adjustment value of the cooling medium to obtain control parameters for the two areas; The control parameters are transmitted to the heating device and the cooling medium adjustment module to perform power adjustment and flow adjustment operations and obtain real-time adjusted temperature data; If the real-time adjusted temperature data exceeds the preset temperature range, the adjustment parameters are recalculated by the control algorithm and transmitted to the relevant modules for secondary adjustment to determine whether the temperature returns to the stable range; If the temperature returns to the stable range, the flow stability data is continuously monitored, and the dynamic change of the fluctuation range is obtained in combination with the historical temperature change trend to determine the fluctuation control effect; By using the fluctuation control effect data, the stable control status of the molten sulfur area and the liquid sulfur area is analyzed, and the prediction results of long-term temperature fluctuations are obtained by using the time series analysis method; According to the prediction result, if the long-term fluctuation range exceeds the preset threshold, the parameter configuration of the control algorithm is adjusted, the power adjustment value and the flow regulation value are recalculated, and the optimized stable control state is determined.

7. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: The method of obtaining adjusted sulfur flow state data based on the fluctuation amplitude, calculating the flow velocity and viscosity changes of sulfur in the pipeline through fluid mechanics simulation, and determining the degree of flow state optimization includes: Extracting initial flow state information from the fluctuation amplitude data of sulfur in the pipeline, and comparing it with a pre-established database to obtain preliminary flow characteristics of the sulfur in the pipeline environment; Based on the preliminary flow characteristics, the flow velocity distribution and viscosity parameter changes of the sulfur in the pipeline are obtained, and multi-dimensional calculations are performed using a fluid mechanics simulation tool to determine the dynamic distribution results of the flow velocity and viscosity; Analyze the influence of the fluctuation amplitude on the flow velocity distribution according to the dynamic distribution result. If the fluctuation amplitude exceeds a preset threshold, adjust the simulation calculation parameters to obtain optimized flow velocity distribution data. Calculating the interaction between viscosity change and pipeline environment based on the optimized flow velocity distribution data; if the viscosity parameter deviates from a preset range, calibrating it using a finite element analysis method to obtain an adjusted viscosity parameter value; By combining the adjusted viscosity parameter value with the fluid mechanics simulation results, the correlation between the flow state and the degree of optimization is analyzed to determine the direction of improving the flow state of sulfur in the pipeline; In view of the improvement direction, the support vector machine algorithm is used to comprehensively predict the fluctuation amplitude, flow velocity distribution and viscosity parameters to determine the final optimization degree data.

8. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: The process optimizes the flow state and extracts real-time indicators of hydrogenation reaction efficiency and production process stability from the process system. The performance changes after sulfur replaces dimethyl disulfide are analyzed through a data fusion algorithm to obtain process suitability evaluation results, including: Acquiring multi-source data during a hydrogenation reaction, the multi-source data including real-time indicators related to flow conditions, and constructing an initial data set; Based on the initial data set, a data fusion algorithm is used to integrate the multi-source data to obtain a comprehensive characteristic value of hydrogenation reaction efficiency and production process stability; Analyzing the performance change trend after sulfur replaces dimethyl disulfide based on the comprehensive characteristic value, and determining whether the performance change trend exceeds a preset optimization degree threshold; If the performance change trend exceeds a preset optimization degree threshold, the corresponding flow state parameters are recorded; Analyze the fluctuation range of the stability index of the production process based on the flow state parameters and the changes in the real-time indicators to determine whether the stability index meets the process system requirements; If the stability index meets the process system requirements, the improvement in hydrogenation reaction efficiency is compared with the results of the data fusion algorithm to obtain a preliminary conclusion on process suitability; Based on the preliminary conclusions, combined with the comprehensive analysis of the performance change trends and the stability indicators, the support vector machine algorithm was used to verify the process suitability evaluation results to obtain the final suitability evaluation data.

9. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: After obtaining the process suitability assessment results, the operating costs and economic benefit improvement of the sulfur replacement solution are calculated in combination with the cost accounting model. By comparing the benchmark cost of dimethyl disulfide, the cost control effect is judged, including: Acquire process suitability evaluation data from a database, and use pre-established data cleaning rules to remove noise and unify the format of the evaluation data to obtain a preliminarily processed evaluation data set; Based on the preliminarily processed evaluation data set, a cost accounting model is used to calculate the operating cost of the sulfur replacement solution and determine the detailed data of each cost component; Based on the cost structure details and in combination with the economic benefit evaluation logic, the economic benefit improvement of the sulfur replacement solution is calculated to obtain the corresponding benefit index value; By comparing the benefit index value with the preset benchmark cost data, it is determined whether the operating cost is lower than the benchmark cost. If it is lower than the benchmark cost, it is marked as cost control effective; If the cost control mark is effective, further extract the key factors affecting the economic benefit improvement, use the regression analysis model to weight the factors, and determine the main cost driving factors; Based on the main cost driving factors, optimized sulfur substitution scheme adjustment parameters are generated, and adjusted operating costs and economic benefit forecast values ​​are output.

10. The method for precise temperature control and flow state optimization of sulfur replacing dimethyl disulfide in a coal tar hydrogenation process according to claim 1, characterized in that: The above-mentioned cost control effect is analyzed by using a time series prediction algorithm to analyze the sustainability of temperature control and flow state optimization in long-term operation, and a comprehensive optimization plan for production process stability and economy is obtained, including: Acquire temperature control and flow state data from the production process from the historical data acquisition system and organize them into a time series format to obtain the initial data set; Based on the initial data set, an autoregressive integrated moving average model in the time series prediction method is used to analyze the changing trends of temperature control and flow state and determine the fluctuation range of key influencing factors; If the fluctuation range of the key influencing factors exceeds the preset threshold, the abnormal values ​​are cleaned by the data filtering module to obtain the cleaned data set and judge the stability of the influencing factors; For the cleaned data set, combined with the cost control index data, the impact of temperature control and flow state on economic efficiency is analyzed to obtain the cost-effectiveness correlation model; According to the cost-effectiveness correlation model, an optimization algorithm is used to adjust the parameters of temperature control and flow state, obtain the adjusted parameter combination, and determine the optimal configuration of the production process; If the optimized configuration meets the stability index during the simulation operation, the optimized configuration is subjected to multiple rounds of testing by the data verification module to obtain the final stability evaluation result; Based on the final stability evaluation result, an execution strategy of the comprehensive optimization solution is generated to obtain a production process adjustment plan suitable for long-term operation.

Images