Dynamic intelligent preheating system for pulverized coal

Abstract

The invention provides a dynamic intelligent preheating method and system for pulverized coal, and relates to the technical field of intelligent management.The dynamic intelligent preheating method comprises the steps that after a heating starting instruction of a human-computer interface is received, a high-effect heat conduction oil temperature rise module is triggered to send a valve opening instruction to a PLC control system; based on the valve opening instruction, a pneumatic ball valve of a heat conduction oil heating main pipe is controlled to be opened; under the state that the pneumatic ball valve is opened, a variable frequency starting instruction of the gear pump and an opening instruction of the pneumatic control valve are triggered after delaying for 5S; based on the real-time pressure, temperature and flow data of the heat conduction oil header pipe after the pneumatic control valve is opened, the opening degree of the pneumatic control valve and the operation frequency of the gear pump frequency converter are dynamically adjusted through a self-adaptive algorithm module; and in the dynamic adjusting process, corresponding working condition data are recorded and stored in real time based on each operation result of the opening degree of the adjusting valve and the frequency of the frequency converter. According to the invention, the energy consumption cost of preheating unit pulverized coal is reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent management and processing technology, specifically to a dynamic intelligent preheating system for pulverized coal. Background Technology

[0002] In industrial production, pulverized coal preheating plays a crucial role in the stability, energy efficiency, and environmental performance of subsequent processes, and is widely used in key scenarios such as coal-fired power generation, steel smelting, and chemical synthesis. With the expansion of industrial production scale and technological upgrades, traditional pulverized coal preheating technology can no longer meet the demands of modern industry for high efficiency, precision, and intelligence, making technological upgrades urgently needed.

[0003] Currently, existing pulverized coal preheating technologies still face many bottlenecks that urgently need to be addressed in practical applications: Firstly, heat transfer efficiency and energy consumption are significant issues. Traditional preheating methods often rely on indirect heat transfer via media such as hot air or steam. Heat transfer efficiency is greatly affected by factors such as media transport losses and limited heat exchange area. Especially when processing high-moisture pulverized coal (moisture content > 15%) or fine-particle pulverized coal (particle size < 50 μm), the long heat conduction path and large heat loss not only increase preheating time and waste energy by more than 20%, but also easily lead to a decline in pulverized coal quality due to localized overheating or uneven heating, affecting subsequent combustion or processing effects. Simultaneously, traditional systems exhibit low energy conversion efficiency during start-up, shutdown, and load adjustments, further exacerbating the energy consumption problem.

[0004] Secondly, the temperature control accuracy and dynamic response are insufficient. In high-precision industrial scenarios, such as the preheating of pulverized coal injected into blast furnaces, the preheating temperature fluctuation must be controlled within ±5℃ to ensure injection effect and combustion efficiency. However, existing preheating systems mostly use fixed-parameter PID control or simple logic control, which has poor adaptability to changes in operating conditions. When parameters such as pulverized coal flow rate, initial temperature, and moisture content undergo sudden changes (e.g., fluctuation amplitude > 10%), the system adjustment lags significantly, with response time often exceeding 30 seconds, resulting in temperature deviations of ±15℃ or more. This not only affects the subsequent processing quality of pulverized coal but may also increase pollutant emissions.

[0005] Third, the system lacks intelligence and adaptive capabilities. The control logic of existing pulverized coal preheating systems is mostly based on preset parameters or human experience, lacking the ability to dynamically perceive and autonomously optimize real-time operating conditions. During system operation, operators must manually adjust equipment parameters based on experience, making it difficult to achieve closed-loop control of "perception-decision-execution." Especially when facing complex and changing production conditions (such as differences in moisture and particle size between different batches of pulverized coal), the system cannot automatically correct its control strategy, resulting in poor preheating stability and difficulty in achieving optimal energy efficiency over the long term. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a dynamic intelligent preheating method and system for pulverized coal, which can improve energy utilization by 15%-20% and reduce the energy consumption cost per unit of pulverized coal preheating.

[0007] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: In a first aspect, a dynamic intelligent preheating method for pulverized coal is provided, the method comprising high-efficiency heat transfer oil heating, intelligent temperature-controlled dynamic circulation preheating of heat transfer oil, and adaptive learning analysis, wherein the high-efficiency heat transfer oil heating includes the following steps: After receiving the heating start command from the human-machine interface, the high-efficiency heat transfer oil temperature rise module is triggered to send a valve opening command to the PLC control system; Based on the valve opening command, the pneumatic ball valve of the heat transfer oil heating main is controlled to open; After a 5-second delay, the variable frequency start command of the gear pump and the opening command of the pneumatic regulating valve are triggered when the pneumatic ball valve is fully opened. Based on the real-time pressure, temperature and flow data of the heat transfer oil main after the pneumatic regulating valve is opened, the opening degree of the pneumatic regulating valve and the operating frequency of the gear pump frequency converter are dynamically adjusted through the adaptive algorithm module. During the dynamic adjustment process, the corresponding operating data is recorded and stored in real time based on the operation results of each operation of the regulating valve opening and the frequency of the frequency converter. Based on the stored operating condition data, the final control valve opening and inverter frequency under steady-state operating conditions are matched through the analysis and self-learning results of the adaptive algorithm module. Based on the daily newly added operating condition data, the adaptive algorithm module is triggered to perform a data self-tuning update operation.

[0008] Furthermore, the high-efficiency heat transfer oil heating process also includes the following steps: After the gear pump starts running, it triggers real-time monitoring by the pressure and temperature sensors on the outside of the heat transfer oil circulation tank based on the flow results of the heat transfer oil being pumped from the heat transfer oil circulation tank. When the temperature or pressure exceeds the set threshold, the safety interlock state is activated. Based on the activation result of the safety interlock state, the expansion valve is controlled to open to release the pressure or air in the heat transfer oil circulation tank. The cooling system is started synchronously, and the temperature of the heat transfer oil circulation box is reduced based on the input of the cooling medium. Based on the continuous activation result of the safety interlock state, the abnormal flashing display on the human-machine interface and the alarm output of the external sound and light alarm are triggered.

[0009] Furthermore, the intelligent temperature-controlled heat transfer oil dynamic circulation preheating includes: After receiving the preheating start command from the human-machine interface, the system determines whether the preheating set value has been reached based on the real-time oil temperature data output by the high-efficiency heat transfer oil heating module. Based on the judgment result that the preheating set value has been reached, the pneumatic ball valve of the heat transfer oil input main is triggered to open; After a 5-second delay, the gear pump frequency converter start command and the pneumatic regulating valve opening command are triggered when the pneumatic ball valve is fully opened. Based on the operating results of the gear pump and pneumatic regulating valve, flow and temperature data are collected in real time through vortex flow meter and temperature sensors at the beginning and end of the preheating tube; Based on the collected flow and temperature data, the opening of the pneumatic regulating valve and the frequency of the gear pump are dynamically controlled to ensure that the heat transfer oil passes through the preheating pipe at a uniform speed. Based on the uniform flow of the heat transfer oil output, the conical oil guide structure of the heat transfer oil distributor is triggered to evenly distribute the heat transfer oil to multiple heat transfer oil input branches.

[0010] Furthermore, the intelligent temperature-controlled heat transfer oil dynamic circulation preheating also includes the following steps: Based on the detection result of the opening status of the pneumatic ball valve of the pulverized coal conveying branch pipe, the opening permission of the pneumatic ball valve of the heat transfer oil input main pipe is triggered. During the dynamic control of the pneumatic regulating valve opening and frequency conversion, the pipeline flow rate, hourly coal quantity of the upstream coal injection system, preheating pipeline temperature, frequency converter frequency and regulating valve opening data are recorded in real time. Based on the recorded multi-source data, the system parameters are self-tuned by triggering the big data analysis results of the adaptive learning module. Based on the results of the self-tuning operation, an intelligent multivariable data unit library is created and updated in real time within the system; Based on the flow results of the heat transfer oil output from the preheating pipeline, the heat transfer oil is sequentially passed through the heat transfer oil output branch pipe, the pneumatic ball valve and the heat transfer oil collector to trigger the closed-loop return flow of the heat transfer oil to the circulation box. Based on the circulation results formed by the closed-loop reflux, the continuous preheating of the pulverized coal branch pipe is maintained.

[0011] Furthermore, the adaptive learning parsing includes: Based on the preset target value of pipeline flow velocity, the vortex flow meter is triggered to measure the actual flow velocity in real time. Based on the difference between the target value and the measured value, a control signal is generated. Based on the processing results of the control signals, the frequency adjustment action of the synchronous drive gear pump inverter and the opening adjustment action of the pneumatic regulating valve are controlled. Based on the execution results of frequency regulation and opening regulation, a new round of actual flow velocity measurement is triggered to form a closed-loop feedback; During the closed-loop control process, based on the continuously collected actual values ​​of pressure, flow, and temperature and their deviations from the target values, combined with the historical operation records of frequency converter commands and regulating valve opening commands, the dynamic model is triggered to be updated online. Based on the updated dynamic model and real-time system status, a combination of frequency and opening degree adjustment instructions that minimizes the cost function is generated through rolling optimization calculations.

[0012] Secondly, a dynamic intelligent preheating system for pulverized coal includes: The receiving module is used to receive the heating start command from the human-machine interface and then trigger the high-efficiency heat transfer oil temperature rise module to send a valve opening command to the PLC control system. The control module is used to control the pneumatic ball valve of the heat transfer oil heating main to open based on the valve opening command; The opening module is used to trigger the frequency conversion start command of the gear pump and the opening command of the pneumatic regulating valve after a 5-second delay when the pneumatic ball valve is fully opened. The adjustment module is used to dynamically adjust the opening degree of the pneumatic control valve and the operating frequency of the gear pump frequency converter based on the real-time pressure, temperature and flow data of the heat transfer oil main after the pneumatic control valve is opened, through an adaptive algorithm module. The storage module is used to record and store the corresponding operating condition data in real time based on the operation results of each operation of the regulating valve opening and the frequency of the frequency converter during the dynamic adjustment process. The matching module is used to match the final control valve opening and inverter frequency under steady-state conditions based on the stored operating condition data and the analysis and self-learning results of the adaptive algorithm module; based on the newly added operating condition data each day, it triggers the data self-tuning update operation of the adaptive algorithm module.

[0013] Thirdly, a computing device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0014] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0015] The above-described solution of the present invention has at least the following beneficial effects: Using heat transfer oil as the heat transfer medium, compared to traditional hot air and steam heating, heat transfer oil has a higher specific heat capacity and thermal conductivity, enabling more efficient heat transfer. Experimental data shows that under the same preheating conditions, this invention can improve energy utilization by 15%-20%, significantly reducing the energy cost per unit of pulverized coal preheating. An adaptive algorithm module analyzes the pressure, temperature, and flow data of the heat transfer oil main in real time, dynamically adjusting the opening of the pneumatic regulating valve and the frequency of the gear pump inverter, ensuring the system maintains optimal operating conditions under different loads. For example, even with pulverized coal flow fluctuations of ±15%, the system can still intelligently control energy consumption fluctuations within ±5%, reducing additional energy consumption by 10%-15% compared to traditional systems.

[0016] Based on self-learning and analysis of historical operating data, the system can automatically match the optimal parameter combination under steady-state conditions. In practical applications, the energy utilization rate during steady-state operation is 8%-12% higher than that during the initial commissioning phase, and during long-term operation, the system can continuously maintain optimal energy efficiency through daily data self-tuning updates.

[0017] By combining real-time data acquisition with adaptive algorithms, the parameters of the heat transfer oil are dynamically adjusted, reducing the response time to less than 5 seconds (compared to >30 seconds for traditional PID control). When handling high-moisture (moisture content >15%) or fine-particle (particle size <50μm) coal powder, the temperature control accuracy reaches ±3℃, an improvement of ±10℃ compared to traditional technologies, effectively avoiding local overheating or uneven heating. By simultaneously adjusting the opening of the pneumatic regulating valve and the frequency of the gear pump, coordinated control of the heat transfer oil flow, pressure, and temperature is achieved. Under complex operating conditions such as sudden changes in coal powder flow (e.g., ±20%) or initial temperature fluctuations (±10℃), the system can still quickly stabilize the preheating temperature within the target range of ±5℃, ensuring the stability of subsequent processes.

[0018] Precise temperature control effectively reduces temperature fluctuations during pulverized coal preheating, keeping the pulverized coal moisture content fluctuation within ±0.5% (compared to ±2% for traditional technology). This significantly improves the stability of pulverized coal quality, providing a reliable guarantee for subsequent combustion or processing stages and reducing the risk of product quality problems and production accidents caused by temperature fluctuations.

[0019] By continuously recording and analyzing operating data, the adaptive algorithm is continuously optimized. In practical applications, the system can learn and self-tune parameters for common operating conditions within 30 days of operation, improving control accuracy by 15%-20%. At the same time, the daily data update mechanism ensures that the system can adapt to changes in coal powder characteristics (such as moisture and particle size fluctuations) and equipment performance degradation, improving long-term operational stability by more than 30% compared to traditional systems.

[0020] Based on in-depth analysis of historical data, the system can predict potential faults (such as valve jamming and pump wear) in advance and issue early warnings before the faults occur, reducing unplanned downtime. Statistics show that after adopting this invention, equipment maintenance costs have decreased by 20%-25%, and the number of unplanned downtimes has decreased by more than 50%.

[0021] It achieves fully automated control from startup to operation, reducing reliance on human experience. Operators only need to set target parameters through the human-machine interface, and the system can automatically complete the control and optimization of the preheating process, reducing operational intensity by more than 60% and reducing the risk of production accidents caused by human error. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of a dynamic intelligent preheating system for pulverized coal. Detailed Implementation

[0023] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0024] like Figure 1 As shown, embodiments of the present invention propose a dynamic intelligent preheating system for pulverized coal. The core objective is to optimize combustion efficiency, reduce energy consumption, and decrease pollutant emissions by dynamically heating pulverized coal to improve its temperature and dryness. This system integrates multiple disciplines, including thermal control and intelligent algorithms, to achieve automation, precision, and high efficiency in the pulverized coal preheating process. Through technological innovation, the dynamic intelligent preheating system for pulverized coal combustion pretreatment achieves intelligent and efficient processing, making it a key piece of equipment for energy conservation, emission reduction, and green transformation in the coal-fired industry. With increasingly stringent environmental requirements and the in-depth application of intelligent technologies, this system is expected to replace traditional preheating methods in more fields, becoming an important support for the clean and efficient utilization of coal. In the future, it is necessary to continuously overcome key technological bottlenecks, improve system reliability and economy, and promote the overall upgrading of the industry.

[0025] The following are the specific implementation steps and plans. The pulverized coal dynamic intelligent preheating system consists of three subsystems: Clicking the "Start Heating" button on the human-machine interface triggers the high-efficiency heat transfer oil temperature rise module to start, which then transmits the start command to the PLC control system.

[0026] After receiving the start command, the PLC control system controls the pneumatic ball valve 32 of the heat transfer oil heating main pipe to open.

[0027] After the pneumatic ball valve 32 of the heat transfer oil heating main pipe is opened, there is a 5-second delay before the PLC control system sends a start command to the frequency converter, and the frequency converter controls the gear pump 33 to start working; at the same time, the PLC control system controls the pneumatic regulating valve 34 to open.

[0028] After the gear pump 33 starts working, it draws heat transfer oil from the heat transfer oil circulation tank 25. The heat transfer oil flows through the pneumatic ball valve 32 and the pneumatic regulating valve 34 of the heat transfer oil heating main pipe in sequence, and then enters the heat transfer oil heating main pipe 36 (iron trough heating area).

[0029] The adaptive algorithm module collects pressure, temperature and flow data in the heating manifold of the heat transfer oil in real time. Based on the set target values ​​of pressure, flow and temperature, it dynamically adjusts the opening of the pneumatic regulating valve 34 and the output frequency of the frequency converter to regulate the flow and pressure of the heat transfer oil.

[0030] The heat transfer oil, after being heated by the heat transfer oil heating main pipe 36 (iron trough heating area), returns to the heat transfer oil circulation tank 25 along the pipeline.

[0031] Pressure sensor 24 and temperature sensor 23 installed on the outside of the heat transfer oil circulation tank 25 monitor the pressure and temperature inside the heat transfer oil circulation tank 25 in real time and transmit the monitoring data to the PLC control system.

[0032] During the adjustment process, the adaptive algorithm module records and stores the opening degree of the pneumatic control valve 34, the frequency of the frequency converter, and the corresponding operating conditions such as pressure, temperature, and flow rate in real time. Through analysis and self-learning of these data, it matches the optimal opening degree of the pneumatic control valve 34 and the frequency converter frequency under the current operating conditions.

[0033] The adaptive algorithm module performs daily self-tuning updates based on newly added operating data, updating the optimal pneumatic control valve 34 opening degree and inverter frequency parameters to ensure the system operates under optimal process parameters.

[0034] When the pressure sensor 24 detects that the pressure in the heat transfer oil circulation tank 25 exceeds the set value, or the temperature sensor 23 detects that the temperature in the heat transfer oil circulation tank 25 exceeds the set value, the PLC control system triggers the safety interlock state.

[0035] Once the safety interlock is activated, the PLC control system automatically controls the expansion valve 29 to open, releasing the pressure or air in the heat transfer oil circulation tank 25; at the same time, the cooling system is activated to accelerate the cooling of the heat transfer oil in the heat transfer oil circulation tank 25.

[0036] Once the safety interlock is triggered, the HMI displays "Abnormal temperature and pressure in the heat transfer oil circulation box" and flashes, while the external audible and visual alarm simultaneously sounds an alarm to prompt the operator to take action.

[0037] 2. Intelligent temperature-controlled dynamic circulation preheating system for heat transfer oil Clicking the "Start Preheating" button on the human-machine interface triggers the start of the intelligent temperature-controlled heat transfer oil dynamic circulation module.

[0038] After the intelligent temperature control heat transfer oil dynamic circulation module is started, it first acquires the heat transfer oil temperature data monitored by the heat transfer oil circulation box temperature sensor 23 in the high-efficiency heat transfer oil heating system, and determines whether the temperature has reached the set value.

[0039] If the temperature of the heat transfer oil reaches the set value, the intelligent temperature control heat transfer oil dynamic circulation module sends an "open" command to the PLC control system. After receiving the command, the PLC control system controls the pneumatic ball valve 18 of the heat transfer oil input main pipe to open.

[0040] After the pneumatic ball valve 18 of the heat transfer oil inlet main is opened, there is a 5-second delay before the PLC control system sends a start command to the frequency converter, and the frequency converter controls the gear pump 16 to start working; at the same time, the PLC control system controls the pneumatic regulating valve 17 to open.

[0041] After the gear pump 16 starts working, it draws heat transfer oil from the heat transfer oil circulation tank 25. The heat transfer oil flows sequentially through the pneumatic ball valve 18 of the heat transfer oil inlet main pipe, the vortex flow meter 15, the pneumatic regulating valve 17, the gear pump 16, and the check valve 20 of the heat transfer oil inlet main pipe before entering the heat transfer oil inlet main pipe 19.

[0042] The vortex flow meter 15 monitors the flow rate data of the heat transfer oil in real time and transmits the data to the intelligent temperature control heat transfer oil dynamic circulation module; at the same time, the temperature sensor 02 at the tail end of the preheating pipe and the temperature sensor 04 at the head end of the preheating pipe monitor the temperature data at the tail end and head end of the preheating pipe 03, respectively, and transmit the data to the module.

[0043] The intelligent temperature-controlled heat transfer oil dynamic circulation module dynamically adjusts the opening degree of the pneumatic regulating valve 17 and the output frequency of the frequency converter based on the flow data of the vortex flow meter 15, the temperature data of the preheating pipe tail end temperature sensor 02 and the preheating pipe head end temperature sensor 04, and the set temperature and flow target values, thereby controlling the flow rate of the heat transfer oil through the preheating pipe and making the heat transfer oil pass through the heat transfer oil distributor 13 at a uniform and slow speed.

[0044] The heat transfer oil is evenly distributed to multiple heat transfer oil inlet branches 55 through the conical oil guide structure inside the heat transfer oil distributor 13. Each heat transfer oil inlet branch 55 is connected to a different preheating pipe 03, ensuring that the heat transfer oil distribution in each preheating pipe 03 is uniform, thereby ensuring a uniform preheating temperature.

[0045] The intelligent temperature-controlled heat transfer oil dynamic circulation module acquires the on / off status data of the pneumatic ball valves 06 of each pulverized coal conveying branch in real time, and controls the corresponding pneumatic ball valve 11 of the heat transfer oil input main to open or close according to the status, so as to realize the coordinated linkage between heat transfer oil supply and pulverized coal conveying.

[0046] Powdered coal is transported to the powdered coal distributor 08 via the main powdered coal conveying pipe 09. The powdered coal distributor 08 distributes the powdered coal to each powdered coal conveying branch pipe 05. The pneumatic ball valve 06 and the manual ball valve 07 of the powdered coal conveying branch pipe are used to control the on / off of the powdered coal conveying in the corresponding powdered coal conveying branch pipe 05. When open, the powdered coal enters the preheating pipe 03 along the powdered coal conveying branch pipe 05.

[0047] During the flow of pulverized coal in the preheating pipe 03, it exchanges heat with the heat transfer oil flowing through the heat transfer oil input branch pipe 55 and surrounding the preheating pipe 03, thus completing the preheating process.

[0048] After participating in heat exchange, the heat transfer oil flows out from the preheating pipe 03, and flows sequentially through the preheating pipe outlet 01, the heat transfer oil output branch pipe 54, the manual ball valve 51 of the first heat transfer oil output branch pipe, the manual ball valve 53 of the second branch pipe, and the pneumatic ball valve 52 of the heat transfer oil output branch pipe, and finally gathers into the heat transfer oil collector 50.

[0049] The heat transfer oil collected by the heat transfer oil collector 50 returns to the heat transfer oil circulation tank 25 through the heat transfer oil output manifold and the heat transfer oil output manifold pneumatic ball valve 37, forming a heat transfer oil circulation.

[0050] The adaptive learning module in the intelligent temperature control heat transfer oil dynamic circulation module receives and records in real time the pipeline flow data monitored by the vortex flow meter 15, the hourly coal quantity data of the coal injection system in the preceding section, the preheating pipeline temperature data monitored by the preheating pipe tail end temperature sensor 02 and the preheating pipe head end temperature sensor 04, the frequency data of the frequency converter, and the opening degree data of the pneumatic regulating valve 17.

[0051] The adaptive learning module analyzes the recorded data and performs automatic tuning through big data analysis and calculation, automatically creating an intelligent multivariate data unit library within the system. At the same time, it updates the multivariate database unit in real time based on the real-time updated data to ensure that the corresponding process parameters can be matched under different operating conditions, thus ensuring the accuracy and timeliness of temperature control.

[0052] After being preheated by preheating pipe 03, the pulverized coal continues to be transported along the pipe and eventually enters the blast furnace tuyeres to ensure complete combustion, improve combustion efficiency, and reduce pollutant emissions.

[0053] Analysis of the Adaptive Learning Module: After the adaptive learning module is started, the variables are defined first: the target value of the pipeline flow velocity is set, and the actual flow velocity of the heat transfer oil in the pipeline is measured in real time by a vortex flow meter as the actual value.

[0054] After receiving the actual value, the adaptive learning module compares it with the set target value and calculates the deviation, which is the difference between the target value and the actual value.

[0055] Based on the calculated deviation magnitude and trend, the adaptive learning module generates a control signal, which is used to drive the actuator to adjust the flow rate.

[0056] The control signal drives the gear pump and pneumatic regulating valve to operate: the gear pump receives the control signal through the frequency converter to change the speed, and the speed change affects the theoretical displacement of the heat transfer oil; the pneumatic regulating valve receives the control signal (4-20mA current signal) through the electrical converter and converts it into a pneumatic pressure signal to change the valve opening, and the change in opening affects the flow resistance and thus changes the flow rate.

[0057] The adaptive learning module initiates the data acquisition process: continuously and frequently acquires process variables, including the actual pressure obtained by the pressure sensor 24 of the heat transfer oil circulation box, the actual flow rate obtained by the vortex flow meter, and the actual temperature obtained by the temperature sensor 23 of the heat transfer oil circulation box, the temperature sensor 02 at the tail end of the preheating pipe, and the temperature sensor 04 at the head end of the preheating pipe; at the same time, it acquires set values ​​(target pressure, target flow rate, target temperature), control outputs (current regulating valve opening command, current frequency converter frequency command), and system status (equipment alarm, operating mode), and stores these data in a time-series database indexed by timestamps, retaining 3-5 years of historical data.

[0058] The collected data is preprocessed as follows: First, Kalman filtering is applied to the raw data to remove measurement noise; then, the data collected by different sensors are aligned according to timestamps; subsequently, abnormal data points caused by sensor failure or communication interruption are identified, removed, or corrected; finally, derived features are calculated based on domain knowledge, including deviations (difference between target pressure and actual pressure, difference between target flow rate and actual flow rate, difference between target temperature and actual temperature) and deviation change rates (pressure deviation change rate, flow rate deviation change rate, temperature deviation change rate).

[0059] Online adaptive computation is performed based on preprocessed data: First, model identification is performed, and the mathematical model describing the process dynamics is updated online using real-time and historical data. The model includes the correlation between actual pressure and control valve opening, inverter frequency, actual flow rate, and actual temperature; the correlation between actual flow rate and control valve opening, inverter frequency, and actual pressure; and the correlation between actual temperature and related variables. Next, rolling optimization is performed. In each control cycle, based on the current state and the latest model, the trajectory of actual pressure, actual flow rate, and actual temperature under different combinations of control valve opening and inverter frequency in the future is predicted. A cost function is defined that includes a tracking error term (the sum of squares of the difference between the target value and the predicted value) and a control quantity change term (the sum of squares of the change in control valve opening and the change in frequency). Under the constraint conditions, the control valve opening and inverter frequency values ​​in the next few control cycles are solved. Only the first step of the control command is executed, and the prediction and optimization are re-predicted in the next cycle.

[0060] Combining reinforcement learning and expert rule-assisted online adjustment: The control process is regarded as a Markov decision process, with deviation, deviation rate of change, valve position, and inverter frequency as states, and the change in valve opening and frequency as actions; at the same time, the "IF-THEN" expert rule base is called, with the rule input being state variables such as deviation and rate of change, and the output being the adjustment direction and magnitude of valve and frequency, and the rule base parameters are automatically adjusted based on historical performance data.

[0061] Perform offline global optimization (daily data self-tuning update): The system automatically selects high-quality, representative operating data from the previous day; it uses rich historical data to train more complex models (due to the large computational load, these are only used for offline optimization). The model aims to accurately predict the optimal control valve opening, inverter steady-state frequency value, and dynamic response characteristics when the system reaches a steady state under given setpoints, environmental conditions, and system load. The system uses historical data in a simulation environment or safety sandbox to verify the new parameter set or strategy obtained from offline optimization, ensuring performance improvement without risk. After successful verification, the new optimal parameter set / strategy is automatically or automatically updated to the online adaptive algorithm module when the system load is low or during planned downtime.

[0062] The system is linked to an industrial-grade liquid cooling temperature control system for safety regulation: When the actual pressure monitored by the pressure sensor 24 of the heat transfer oil circulation tank or the actual temperature monitored by the temperature sensor 23 of the heat transfer oil circulation tank exceeds the set value, the adaptive learning module triggers a safety control signal; the safety control signal drives the pneumatic ball valves 43 and 45 of the cooling water pipe to open, and at the same time starts the slurry pump 44; the slurry pump 44 delivers the cooling water in the circulation water tank 47 to the water distribution nozzle 48 through the cooling water pipe 40, the metal flexible connection 42 of the cooling water pipe, the manual ball valve 41 of the cooling water pipe, and the manual ball valve 46 of the cooling water pipe, and the water distribution nozzle 48 cools the heat transfer oil circulation tank 25; at the same time, the expansion valve 29 opens to release pressure until the actual pressure and temperature drop to within the set value range, and the system exits the safety regulation state.

[0063] The following is a functional description and logic control principle: After the industrial-grade liquid-cooled temperature control module is started, it enters standby mode and establishes a safety interlock with the intelligent temperature-controlled thermal oil dynamic circulation preheating system and the high-efficiency thermal oil heating system, receiving real-time operating status signals from the associated systems.

[0064] The temperature sensor 23 of the heat transfer oil circulation box monitors the temperature data inside the heat transfer oil circulation box 25 in real time, and the pressure sensor 24 of the heat transfer oil circulation box monitors the pressure data inside the heat transfer oil circulation box 25 in real time. Both types of data are transmitted synchronously to the control unit of the industrial-grade liquid cooling temperature control module.

[0065] After receiving temperature and pressure data, the control unit of the industrial-grade liquid-cooled temperature control module compares the actual temperature with the set safe temperature value and the actual pressure with the set safe pressure value to determine whether the actual temperature and pressure are both greater than or equal to the set safe value.

[0066] If the actual temperature and pressure are both greater than or equal to the set safety value, the control unit immediately issues a control command to drive the expansion valve 29 to open automatically, and at the same time controls the opening of the cooling water pipe pneumatic ball valve 43 and the cooling water pipe pneumatic ball valve 45, and starts the slurry pump 44.

[0067] After the slurry pump 44 starts, it draws out the cooling water from the circulating water tank 47. The cooling water is then transported to the water distribution nozzle 48 via the cooling water pipe 40. The water distribution nozzle 48 sprays the cooling water into the heat transfer oil circulating tank 25 to cool it down. The cooled water that has absorbed heat flows back to the circulating water tank 47 to form a cooling cycle, which reduces the temperature and pressure inside the heat transfer oil circulating tank 25.

[0068] The control unit synchronously triggers the alarm mechanism, and the human-machine interface (HMI) displays the alarm message "abnormal temperature and pressure of heat transfer oil circulation box" and flashes continuously. The external audible and visual alarm simultaneously issues an audible and visual alarm to prompt the operator to pay attention to the system abnormality.

[0069] The temperature sensor 23 and pressure sensor 24 of the heat transfer oil circulation box continuously monitor the temperature and pressure changes inside the heat transfer oil circulation box 25 and feed the data back to the control unit in real time.

[0070] The control unit continuously receives updated temperature and pressure data, compares the actual temperature with the set safe temperature value, and compares the actual pressure with the set safe pressure value to determine whether the actual temperature and pressure are both less than or equal to the set safe value.

[0071] When the actual temperature and pressure are both less than or equal to the set safety value, the control unit issues a stop command, first controlling the slurry pump 44 to stop running, and then sequentially closing the cooling water pipe pneumatic ball valve 43 and the cooling water pipe pneumatic ball valve 45. The expansion valve 29 is also closed, terminating the cooling cycle and pressure release.

[0072] The adaptive learning module in the industrial-grade liquid cooling temperature control module records complete data for each abnormal event, including the time of occurrence, peak temperature and pressure, duration, and cooling system operating parameters, and stores this data in a database.

[0073] The adaptive learning module periodically summarizes and analyzes historical alarm data stored in the database, and uses big data analysis algorithms to mine the patterns and correlations of temperature and pressure changes, and calculates the temperature warning value and pressure warning value for the corresponding operating conditions.

[0074] The system presets the calculated early warning value into the monitoring logic. During subsequent operation, when the temperature or pressure of the heat transfer oil circulation box 25 reaches the early warning value, it will trigger an early warning prompt in advance, thereby realizing the prediction of abnormal risks in the system and avoiding production safety accidents.

[0075] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A dynamic intelligent preheating method for pulverized coal, characterized in that, The method includes high-efficiency heat transfer oil heating, intelligent temperature-controlled dynamic circulation preheating of heat transfer oil, and adaptive learning analysis. The high-efficiency heat transfer oil heating includes the following steps: After receiving the heating start command from the human-machine interface, the high-efficiency heat transfer oil temperature rise module is triggered to send a valve opening command to the PLC control system; Based on the valve opening command, the pneumatic ball valve of the heat transfer oil heating main is controlled to open; After a 5-second delay, the variable frequency start command of the gear pump and the opening command of the pneumatic regulating valve are triggered when the pneumatic ball valve is fully opened. Based on the real-time pressure, temperature and flow data of the heat transfer oil main after the pneumatic regulating valve is opened, the opening degree of the pneumatic regulating valve and the operating frequency of the gear pump frequency converter are dynamically adjusted through the adaptive algorithm module. During the dynamic adjustment process, the corresponding operating data is recorded and stored in real time based on the operation results of each operation of the regulating valve opening and the frequency of the frequency converter. Based on the stored operating condition data, the final control valve opening and inverter frequency under steady-state operating conditions are matched through the analysis and self-learning results of the adaptive algorithm module. Based on the daily newly added operating condition data, the adaptive algorithm module is triggered to perform a data self-tuning update operation.

2. The method for dynamic intelligent preheating of pulverized coal according to claim 1, characterized in that, The high-efficiency heat transfer oil heating process also includes the following steps: After the gear pump starts running, it triggers real-time monitoring by the pressure and temperature sensors on the outside of the heat transfer oil circulation tank based on the flow results of the heat transfer oil being pumped from the heat transfer oil circulation tank. When the temperature or pressure exceeds the set threshold, the safety interlock state is activated. Based on the activation result of the safety interlock state, the expansion valve is controlled to open to release the pressure or air in the heat transfer oil circulation tank. The cooling system is started synchronously, and the temperature of the heat transfer oil circulation box is reduced based on the input of the cooling medium. Based on the continuous activation result of the safety interlock state, the abnormal flashing display on the human-machine interface and the alarm output of the external sound and light alarm are triggered.

3. The method for dynamic intelligent preheating of pulverized coal according to claim 2, characterized in that, The intelligent temperature-controlled heat transfer oil dynamic circulation preheating includes: After receiving the preheating start command from the human-machine interface, the system determines whether the preheating set value has been reached based on the real-time oil temperature data output by the high-efficiency heat transfer oil heating module. Based on the judgment result that the preheating set value has been reached, the pneumatic ball valve of the heat transfer oil input main is triggered to open; After a 5-second delay, the gear pump frequency converter start command and the pneumatic regulating valve opening command are triggered when the pneumatic ball valve is fully opened. Based on the operating results of the gear pump and pneumatic regulating valve, flow and temperature data are collected in real time through vortex flow meter and temperature sensors at the beginning and end of the preheating tube; Based on the collected flow and temperature data, the opening of the pneumatic regulating valve and the frequency of the gear pump are dynamically controlled to ensure that the heat transfer oil passes through the preheating pipe at a uniform speed. Based on the uniform flow of the heat transfer oil output, the conical oil guide structure of the heat transfer oil distributor is triggered to evenly distribute the heat transfer oil to multiple heat transfer oil input branches.

4. The method for dynamic intelligent preheating of pulverized coal according to claim 3, characterized in that, The intelligent temperature-controlled dynamic circulation preheating of heat transfer oil also includes the following steps: Based on the detection result of the opening status of the pneumatic ball valve of the pulverized coal conveying branch pipe, the opening permission of the pneumatic ball valve of the heat transfer oil input main pipe is triggered. During the dynamic control of the pneumatic regulating valve opening and frequency conversion, the pipeline flow rate, hourly coal quantity of the upstream coal injection system, preheating pipeline temperature, frequency converter frequency and regulating valve opening data are recorded in real time. Based on the recorded multi-source data, the system parameters are self-tuned by triggering the big data analysis results of the adaptive learning module. Based on the results of the self-tuning operation, an intelligent multivariable data unit library is created and updated in real time within the system; Based on the flow results of the heat transfer oil output from the preheating pipeline, the heat transfer oil is sequentially passed through the heat transfer oil output branch pipe, the pneumatic ball valve and the heat transfer oil collector to trigger the closed-loop return flow of the heat transfer oil to the circulation box. Based on the circulation results formed by the closed-loop reflux, the continuous preheating of the pulverized coal branch pipe is maintained.

5. The method for dynamic intelligent preheating of pulverized coal according to claim 4, characterized in that, The adaptive learning parsing includes: Based on the preset target value of pipeline flow velocity, the vortex flow meter is triggered to measure the actual flow velocity in real time. Based on the difference between the target value and the measured value, a control signal is generated. Based on the processing results of the control signals, the frequency adjustment action of the synchronous drive gear pump inverter and the opening adjustment action of the pneumatic regulating valve are controlled. Based on the execution results of frequency regulation and opening regulation, a new round of actual flow velocity measurement is triggered to form a closed-loop feedback; During the closed-loop control process, based on the continuously collected actual values ​​of pressure, flow, and temperature and their deviations from the target values, combined with the historical operation records of frequency converter commands and regulating valve opening commands, the dynamic model is triggered to be updated online. Based on the updated dynamic model and real-time system status, a combination of frequency and opening degree adjustment instructions that minimizes the cost function is generated through rolling optimization calculations.

6. A dynamic intelligent preheating system for pulverized coal, wherein the system implements the method as described in any one of claims 1 to 5, characterized in that, include: The receiving module is used to receive the heating start command from the human-machine interface and then trigger the high-efficiency heat transfer oil temperature rise module to send a valve opening command to the PLC control system. The control module is used to control the pneumatic ball valve of the heat transfer oil heating main to open based on the valve opening command; The opening module is used to trigger the frequency conversion start command of the gear pump and the opening command of the pneumatic regulating valve after a 5-second delay when the pneumatic ball valve is fully opened. The adjustment module is used to dynamically adjust the opening degree of the pneumatic control valve and the operating frequency of the gear pump frequency converter based on the real-time pressure, temperature and flow data of the heat transfer oil main after the pneumatic control valve is opened, through an adaptive algorithm module. The storage module is used to record and store the corresponding operating condition data in real time based on the operation results of each operation of the regulating valve opening and the frequency of the frequency converter during the dynamic adjustment process. The matching module is used to match the final control valve opening and inverter frequency under steady-state conditions based on the stored operating condition data and the analysis and self-learning results of the adaptive algorithm module; based on the newly added operating condition data each day, it triggers the data self-tuning update operation of the adaptive algorithm module.

7. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 5.

Images