A coking heat consumption prediction method, system, device and medium

Abstract

This invention provides a method, system, equipment, and medium for predicting coking calorific value. The method includes: acquiring information on the composition of the coal fed into the furnace, coal charging parameters, historical data on coking calorific value, and preset values ​​of parameters in the coking process parameters; calculating the amount of coal charged through the orifice based on the coal charging parameters; calculating the calorific value affected by 1% of the orifice based on the preset values ​​of the parameters in the coking process parameters; and inputting the information on the composition of the coal fed into the furnace, the coal charging parameters, the historical data on coking calorific value, the preset values ​​of the parameters, the amount of coal charged through the orifice, and the calorific value affected by 1% of the orifice into a pre-trained process mechanism model to predict the coking calorific value. This invention significantly improves the level of refined management of coke oven heating and effectively controls coking production costs.

Description

Technical Field

[0001] This invention relates to the field of data processing and intelligent prediction technology in the coking industry, and in particular to a method, system, equipment and medium for predicting coking calorific value. Background Technology

[0002] The coking industry plays a crucial role in the steel industry chain and is an indispensable link in steel production. As a bridge connecting the coal and steel industries, the coke oven, as the core conversion equipment, primarily transforms coal into coke through high-temperature processing, providing the main fuel and reducing agent for blast furnace ironmaking. In this process, energy consumption becomes a critical evaluation indicator, directly impacting the cost and environmental impact of coking production. Coking calorific value is an important energy consumption quota standard in the coking process. It serves not only as a comprehensive evaluation standard for improving coke oven structure, standardizing temperature control technology, and enhancing coke oven management, but also as a vital basis for optimizing coke oven operation and improving energy utilization efficiency.

[0003] However, due to the complexity of the coking process and the limitations of existing technology, the prediction of coking calorific value relies mainly on human experience for assessment and judgment, making it difficult to meet the actual production requirements in terms of control precision. Summary of the Invention

[0004] This invention provides a method, system, equipment, and medium for predicting coking calorific value, in order to solve the problem that the prediction of coking calorific value in the prior art relies on manual experience assessment or post-event monitoring, which makes it impossible to accurately quantify and evaluate the energy consumption in the coking process, and thus makes it difficult to effectively control the coking production cost.

[0005] This invention provides a method for predicting coking calorific value, comprising: acquiring information on the composition of coal fed into the furnace, information on coal charging parameters, historical data on coking calorific value, and set values ​​of preset parameters in the coking process parameters; calculating the amount of coal charged through the orifice based on the coal charging parameter information; calculating the calorific value affected by 1% of the orifice based on the preset parameters in the coking process parameters; and inputting the information on the composition of coal fed into the furnace, the information on coal charging parameters, the historical data on coking calorific value, the set values ​​of the preset parameters, the amount of coal charged through the orifice, and the calorific value affected by 1% of the orifice into a pre-trained process mechanism model to predict the coking calorific value.

[0006] In one embodiment of the present invention, the coal composition information includes the moisture content and volatile matter content of the coal. The coal loading parameter information includes the total coal loading amount and the number of furnaces loaded; the historical data of coking calorific value includes the lowest coking calorific value within a preset time period.

[0007] In one embodiment of the present invention, the preset parameters in the coking process parameters include the standard value of coking calorific value, the standard coal loading amount of a single-hole carbonization chamber, the moisture content of standard wet coal, the volatile matter content of standard wet coal, the effect of 1% moisture content of coal fed into the furnace on calorific value, and the effect of 1% volatile matter content of coal fed into the furnace on calorific value.

[0008] In one embodiment of the present invention, the step of calculating the coal loading amount of the hole based on the coal loading parameter information includes: calculating the coal loading amount of the hole based on the total coal loading amount and the number of furnaces loaded within a preset time period.

[0009] In one embodiment of the present invention, the step of calculating the 1% hole-affected heat consumption based on the preset parameter setting value includes: calculating the 1% hole-affected heat consumption based on the standard value of coking heat consumption and the standard coal loading amount of a single-hole carbonization chamber within a preset time period.

[0010] In one embodiment of the present invention, the coal composition information, coal charging parameter information, historical data of coking calorific value, preset parameter settings, orifice coal charging amount, and the calorific value affected by 1% of orifices are input into a pre-trained process mechanism model to predict the coking calorific value, including: Based on the aforementioned process mechanism model, the coking calorific value is calculated using the following parameters: coal charge per hole, moisture content of coal fed into the furnace, volatile matter content of coal fed into the furnace, moisture content of standard wet coal, volatile matter content of standard wet coal, standard coal charge per single-hole carbonization chamber, minimum coking calorific value within a preset time period, calorific value affected by 1% moisture content of coal fed into the furnace, calorific value affected by 1% volatile matter content of coal fed into the furnace, and calorific value affected by 1% of the hole.

[0011] In one embodiment of the present invention, after predicting the coking calorific value, the method further includes: comparing the predicted coking calorific value with the target value of the coking calorific value to obtain the fluctuation monitoring result of the predicted coking calorific value; wherein the target value of the coking calorific value is determined by the average value of the predicted coking calorific value within a preset time period.

[0012] The present invention also provides a coking calorific value prediction system, comprising: The data processing module is used to acquire information on the composition of coal fed into the furnace, coal charging parameters, historical data on coking calorific value, and the set values ​​of preset parameters in the coking process parameters. The parameter characteristic value calculation module is used to calculate the amount of coal loaded into the hole based on the coal loading parameter information, and to calculate the heat consumption affected by 1% of the hole based on the set value of the preset parameter. The objective function value prediction module is used to input the coal composition information, coal charging parameter information, historical data of coking calorific value, preset parameter settings, orifice coal charging amount, and the calorific value affected by 1% of the orifices into a pre-trained process mechanism model to predict the coking calorific value.

[0013] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the coking calorific value prediction method.

[0014] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the coking calorific value prediction method.

[0015] The beneficial effects of this invention are as follows: The coking calorific value prediction method, system, equipment, and medium proposed in this invention comprehensively acquire information on the composition of the coal fed into the furnace, coal charging parameters, and historical data on coking calorific value. Combined with preset parameter values ​​in the coking process, it accurately calculates the amount of coal charged into each hole and the calorific value affected by 1% of the holes. By incorporating a deep process mechanism model, it performs multi-dimensional nonlinear fusion calculations of dynamic production data and static benchmark parameters, achieving a leap from empirical fuzzy judgment to scientific quantitative prediction. This feedforward prediction mechanism allows coking oven technicians to anticipate future heat demand over a period of time, thereby enabling them to adjust the heating regime more effectively, greatly improving the level of refined management of coking oven heating, and effectively controlling coking production costs. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0017] In the attached diagram: Figure 1 This is a flowchart illustrating a method for predicting coking heat consumption in one embodiment of the present invention. Figure 2 This is a flowchart illustrating the method for predicting coking heat consumption in another embodiment of the present invention. Figure 3 This is a schematic diagram of information transmission in a method for predicting coking calorific value according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the coking calorific value prediction system in one embodiment of the present invention. Detailed Implementation

[0018] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0019] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. The drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0020] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0021] To enable those skilled in the art to better understand the technical solutions of this invention, before specifically introducing the technical solutions, the key custom terms and physical quantities involved in this specification will be defined and explained in a concentrated manner.

[0022] Moisture content of coal entering furnace Moisture content refers to the percentage of moisture in the coal actually charged into the coke oven. Moisture absorbs latent heat during high-temperature vaporization and is a dynamic variable affecting the calorific value of coking.

[0023] Volatile matter in coal fed into the furnace : refers to the yield of gaseous and liquid products that escape when coal is heated in the furnace at a specific high temperature in the absence of air. The heat released by the partial combustion of these products in the carbonization chamber serves as a dynamic variable for heat consumption compensation.

[0024] Minimum coking heat consumption within the preset time period This refers to the lowest energy consumption baseline value recorded by the coke oven in actual operation within the production cycle closest to the current prediction node. It represents the limit energy efficiency level of the coke oven under the optimal operating conditions in the near future. The preset time period can be the past 30 days, or it can be configured and adjusted according to the actual production cycle.

[0025] Standard value of coking heat consumption : Refers to the baseline theoretical heat energy required for the dry distillation of a unit mass of standard coal under design specifications or calibrated operating conditions.

[0026] Standard coal loading capacity for a single-hole carbonization chamber : Refers to the mass of reference wet coal that a single-hole carbonization chamber can hold under the designed furnace loading density.

[0027] Coal loading capacity : refers to the average actual coal loading value per hole, which is extracted by integrating and summing the total coal loading amount and the number of furnaces loaded within a preset time period. 1% of the pore size affects heat loss The effect of 1% on heat consumption per hole refers to the absolute change in heat consumption caused by a 1% change in the amount of coal charged per hole relative to the standard value.

[0028] Standard wet coal moisture content : Refers to the coal-water distribution ratio used as a design benchmark.

[0029] Standard wet coal volatile matter : Refers to the volatile matter distribution ratio of the blended coal used as a design benchmark.

[0030] The 1% moisture content of the coal fed into the furnace affects the calorific value. : Refers to the correction factor for heat consumption caused by a 1% deviation of the moisture content of the coal entering the furnace from the standard value. The volatile matter content of 1% of the coal fed into the furnace affects the calorific value. : Refers to the correction factor for heat consumption caused by a 1% deviation of the volatile matter content of the coal entering the furnace from the standard value.

[0031] For coking heating systems in coke ovens, energy consumption is a core indicator for assessing production costs. In traditional coking production, due to the diverse materials, complex phase change mechanisms, and strong heat transfer lag in the coking process, the prediction of coking calorific value relies almost entirely on subjective assessments based on manual operational experience. This approach makes it impossible to accurately quantify the actual energy demand during the coking process, and operators often rely on lagging oven temperature measurements for feedback adjustments. This feedback adjustment mechanism suffers from severe thermal inertia lag, leading to irregular fluctuations in gas consumption throughout the production cycle. This not only wastes energy but also easily causes underburning or over-burning of coke, reducing its metallurgical quality. To overcome these technical deficiencies, this invention provides a method, system, equipment, and medium for predicting coking calorific value. By establishing a complete technical closed loop from front-end data acquisition and mid-stage mechanism feature extraction to end-stage objective function fusion calculation, the complex coking mechanism is transformed into a computable mathematical model, achieving a leap from empirical fuzzy judgment to scientific quantitative prediction.

[0032] Please see Figure 1 In some embodiments, the present invention provides a method for predicting coking calorific value to solve the technical problem of the inability to accurately quantify and assess coking energy consumption. The method includes the following steps: Step S100: Obtain information on the composition of coal entering the furnace, information on coal charging parameters, historical data on coking calorific value, and the set values ​​of preset parameters in the coking process parameters.

[0033] In one embodiment, process parameter data during the coking production process can be automatically collected via a PLC (Programmable Logic Controller), a testing system, or other external systems such as L1. Please refer to [link to relevant documentation]. Figure 2 In some embodiments, the coal composition information includes the moisture content of the coal fed into the furnace. Volatile matter of coal fed into the furnace Coal loading parameter information includes total coal loading volume. and the number of furnaces installed Historical data on coking calorific value includes the lowest coking calorific value over the past 30 days. To address the shortcomings of traditional methods, such as limited data collection dimensions and insufficient representativeness, this embodiment comprehensively covers the material flow characteristics and energy flow benchmarks affecting coking calorific value through the synergistic introduction of multi-dimensional data. From a physicochemical reaction mechanism perspective, moisture carries away a significant amount of latent heat during high-temperature vaporization, leading to a decrease in furnace temperature; volatiles partially participate in combustion during dry distillation, releasing heat energy. By precisely introducing these variables, the process mechanism model can fully consider the dynamic balance between heat absorption and release. Simultaneously, it provides the lowest coking calorific value over approximately 30 days. As a historical anchor, it effectively filters out abnormal energy consumption data caused by occasional operational errors in daily production, guides the prediction model to approach the direction of optimal energy efficiency, and improves the robustness of the model in a variable industrial environment.

[0034] In some embodiments, the preset parameters in the coking process parameters include the standard value of coking calorific value. Standard coal loading capacity for a single-hole carbonization chamber Standard wet coal moisture content Standard wet coal volatile matter 1% moisture content in the coal fed into the furnace affects calorific value. The impact of 1% volatile matter in the coal fed into the furnace on calorific value To address the technical challenge of combining static process setpoints with dynamic production data, this embodiment establishes a scientific static benchmark system. This is achieved by introducing two key sensitivity coefficients: and , to the actual moisture and volatile matter The fluctuations deviating from the standard value are linearly mapped to the absolute increase or decrease in heat consumption. This mechanism creates a physically interpretable bias correction engine within the prediction model, which not only reduces the model's dependence on massive data samples during training but also makes the thermodynamic logic of the prediction results clear and traceable, facilitating attribution analysis by technical personnel for fluctuations in specific variables.

[0035] In some embodiments, to more accurately define the statistical significance of dynamic production data, the moisture content of the coal fed into the furnace is... Volatile matter of coal fed into the furnace All values ​​represent the average of the daily test results. Raw coal in industrial settings exhibits spatial and temporal compositional inhomogeneity, and test results from a single sample are prone to significant random errors. This embodiment employs the arithmetic mean of multiple daily test results for smoothing, effectively neutralizing the random bias of a single sample, accurately reflecting the overall physicochemical properties of the coal fed into the furnace that day, and ensuring the high stability and representativeness of the model input data.

[0036] Optionally, since the raw data directly collected by sensors in industrial sites is often mixed with electromagnetic interference, mechanical vibration noise, and transmission glitches, leading to distortion of the prediction model input, this embodiment, after acquiring information on the composition of the coal entering the furnace, coal charging parameters, and historical data on coking calorific value, further includes: for real-time data under different operating conditions, integrating different intelligent methods into the filtering module for data preprocessing; if the production process has square wave-like characteristics, using the moving average filtering method for data preprocessing; if the operating conditions have continuous production characteristics, using the empirical mode decomposition method for data preprocessing.

[0037] By employing the aforementioned hybrid filtering architecture, the shortcomings of single filtering algorithms in adapting to complex industrial environments are overcome. Specifically, for step-like square wave signals such as those generated by coal charging, the moving average filtering can weaken edge overshoot and glitches by setting a sliding window on the time axis. For signals such as those with continuous, slow drift in coal moisture or furnace temperature accompanied by high-frequency white noise, the empirical mode decomposition method can adaptively decompose complex signals into multiple intrinsic mode functions, removing high-frequency noise components and retaining low-frequency trend information. This targeted preprocessing logic significantly improves the signal-to-noise ratio of the input prediction model, providing a high-purity data foundation for subsequent mechanism calculations.

[0038] Step S110: Calculate the coal loading amount in the hole based on the coal loading parameter information.

[0039] In some embodiments, the amount of coal charged in the borehole is calculated based on the coal charging parameter information. This includes: based on a preset time period Total coal loading capacity and the number of furnaces installed Calculate the amount of coal loaded into the hole. The calculation method is as follows:

[0040] in, The amount of coal loaded into the hole. Indicates the first Total coal loading volume at any given time , Indicates the first Number of furnaces loaded at any given time .

[0041] In actual production sites, mechanical errors and differences in operating habits in the coal tower feeding system can lead to significant random fluctuations in the amount of coal charged per operation. If instantaneous data from a single operation is directly adopted, the predicted calorific value curve will exhibit high-frequency oscillations. Using the above equation (1) within the time window... The process involves integral summation and averaging, which is equivalent to building a low-pass filter in the data stream. This effectively smooths out the high-frequency random errors caused by a single coal charging operation, extracts the true macroscopic average single-hole physical load of the coke oven during that time period, and ensures the high fidelity of the source input data.

[0042] Step S120: Calculate the heat loss due to 1% pore size based on the preset parameter settings.

[0043] In some embodiments, the 1% orifice effect heat consumption is calculated based on preset parameters in the coking process parameters. This includes: based on a preset time period Standard value of coking heat consumption in the interior Standard coal loading capacity for single-hole carbonization chamber Calculate the effect of 1% pore size on heat loss. The calculation method is as follows:

[0044] in, The 1% pore size affects heat loss. This is the standard value for coking calorific value. This refers to the standard coal loading amount for a single-hole carbonization chamber.

[0045] Through equation (2) above, this scheme transforms the macroscopic process standard into a microscopic coal loading sensitivity coefficient. During coke oven operation, due to the geometric constraints of the carbonization chamber and the nonlinear changes in coal bulk density, an increase in coal loading is not absolutely proportional to an increase in heat consumption. The derivative relationships between the reference parameters were established, and the complex physical heat transfer process was abstracted into sensitivity coefficients. In subsequent calculations, the thermal correction value can be accurately obtained by multiplying the deviation of the coal loading amount in the hole by this coefficient, which reduces the system's computing power and ensures the accuracy of engineering applications.

[0046] In step S130, the coal composition information, coal charging parameter information, historical data of coking calorific value, preset parameter settings, coal charging amount per hole, and the calorific value affected by 1% of the holes are input into the pre-trained process mechanism model to predict the coking calorific value.

[0047] In some embodiments, the coal composition information, coal charging parameter information, historical data of coking calorific value, preset parameter settings, orifice coal charging amount, and the calorific value affected by 1% of the orifice are input into a pre-trained process mechanism model to predict the coking calorific value Q. This includes: combining the process mechanism model prediction algorithm with the orifice coal charging amount... , the moisture content of the coal entering the furnace Volatile matter in coal fed into the furnace Standard wet coal moisture content Standard wet coal volatile matter Standard coal loading capacity for a single-hole carbonization chamber The lowest coking calorific value in the past 30 days 1% moisture content in the coal fed into the furnace affects calorific value. The volatile matter content of 1% of the coal fed into the furnace affects the calorific value. And 1% pore size affects heat loss Calculate the heat consumption of coking The calculation method is as follows:

[0048] in, This is the predicted target value for coking heat consumption.

[0049] Equation (3) above constitutes the core prediction algorithm engine of this scheme. This engine is not a simple data regression fitting, but has a rigorous physical causal logic: based on the lowest coking calorific value in the past 30 days. As a foundation, dynamic compensation is superimposed across three dimensions. Specifically, this includes mass load compensation (through...). and Comparison of binding coefficients Corrections were made), and moisture heat absorption compensation was achieved (through...). and Comparison of binding coefficients Compensation for latent heat of vaporization of moisture), and compensation for exothermic reaction of volatiles (through... and Comparison of binding coefficients (Deducting the heat replenished by the combustion of internal volatiles). This process mechanism model unifies heterogeneous physical parameters into the heat dimension, realizing a dimensionality reduction mapping of complex thermal systems, solving the problem that multidimensional variables cannot be directly coupled for calculation, and outputting accurate predicted values ​​for coking heat consumption.

[0050] In some embodiments, after predicting coking calorific value, the method further includes: automatically tracking and comparing the fluctuation trends of the daily predicted and actual coking calorific value. The data output by the prediction model needs to be transformed into production management tools, using visual charts to compare the predicted values ​​with the actual calorific value consumed by the coke oven simultaneously on the same screen. By observing the degree of agreement or divergence between the two curves, technicians can quickly verify the effectiveness of the current heating system, identify potential abnormal operating conditions such as furnace air leakage or sudden changes in gas calorific value, and establish a closed-loop management system from algorithm prediction to production intervention.

[0051] Optionally, to enable the prediction system to automatically diagnose abnormal operating conditions, after predicting coking calorific value, the method further includes: comparing the predicted coking calorific value with the target value to obtain the fluctuation monitoring result of the predicted coking calorific value; the target value of coking calorific value is determined by the average value of the predicted coking calorific value within a preset time period. Compared with the shortcomings of traditional alarms using fixed thresholds, which are prone to false alarms and missed alarms, this solution uses a dynamically generated average value as the baseline of the target value, comparing the instantaneous predicted value of the day with the dynamic baseline with time inertia. If the fluctuation difference exceeds the preset tolerance, it can adaptively determine that there is a risk of sudden change in raw material properties or process runaway in the current coke oven, thus improving the high sensitivity identification capability of abnormal operating conditions.

[0052] To address the issue that software-level predictive data cannot directly guide the closed-loop control of the underlying actuators, after predicting the coking calorific value, the method also includes: calculating the corresponding gas flow rate setpoint based on the difference between the predicted coking calorific value and the standard temperature, combined with the gas calorific value; and controlling the opening degree of the gas regulating valve based on the gas flow rate setpoint.

[0053] By transforming abstract heat prediction values ​​into concrete physical execution commands, this embodiment constructs a feedforward-based low-level hardware closed-loop control system. The system calculates the difference between the predicted heat consumption and the theoretical baseline heat under standard operating conditions, combines this with the calorific value parameters of the currently supplied gas, and obtains the gas flow setpoint through thermodynamic conversion. This setpoint is then converted into an electrical signal to directly drive the gas regulating valve. This mechanism overcomes the drawbacks of traditional methods that rely on passively adjusting valves after furnace temperature deviations, resulting in significant thermal inertia lag. It achieves a control breakthrough of "early perception of energy demand and proactive adaptation of gas supply," ensuring the stability of the coke oven thermal regime.

[0054] In some embodiments, to ensure the efficiency of acquiring and the persistence and security of underlying multi-source heterogeneous data, the automatic acquisition of coal composition information and coal charging parameter information, as well as the collection of historical coking calorific value data, includes: automatically collecting process parameter data during the coking production process through a programmable logic controller, a testing and analysis system, or an L1 external system, and automatically saving all parameters to a database. The database may include any one of ORACLE, DB2, SQL Server, Sybase, Informix, MySQL, VF, and Access.

[0055] By establishing a closed-loop technology system from data acquisition and parameter setting to intermediate feature calculation and final target prediction, this embodiment provides pre-emptive, quantifiable data support for energy management in the coking process. It predicts calorific values ​​that would otherwise appear after the fact, providing a clear basis for adjusting the heating regime. By clearly defining the composition information of the coal entering the furnace, such as moisture and volatile matter, as well as charging parameters such as total coal quantity and number of furnaces, it comprehensively covers the material and energy flows affecting coking calorific value. Introducing the lowest coking calorific value within a preset time period as a historical benchmark effectively filters out abnormally high energy consumption data in daily production, ensuring the prediction model consistently converges towards optimal energy efficiency and significantly enhancing its robustness and accuracy in complex and variable industrial environments. Two sensitivity coefficients—one for the 1% moisture content of the coal entering the furnace and one for the 1% volatile matter content—are introduced, and a deviation correction engine is built within the prediction model. When actual production data is input, the model can rationally calculate the deviation between the actual and standard values ​​and convert the deviation into a calorific value using the sensitivity coefficients. This design reduces the complexity of model training, endows the prediction results with strong physical interpretability, and provides a solid theoretical basis for precise temperature control. An automatic tracking and comparison display mechanism is introduced, synchronously comparing the predicted values ​​with actual heat consumption data. This intuitive trend tracking mechanism provides technicians with a review tool, assisting them in making more effective and rigorous production decisions.

[0056] Please see Figure 3The diagram illustrates the hardware-based interactive information transmission architecture of this invention. The coal tower directly supplies raw materials to the coke oven and simultaneously sends coal quantity data to the prediction system. The inspection and testing system extracts and processes samples from the coking process, then sends the coal composition data to the prediction system. The prediction system aggregates multi-source data (such as historical values ​​of coke oven unit consumption and preset parameter values) through an automated link and distributes the processing results and basic data to a designated database for storage. Through hardware-level direct data acquisition technology (such as PLC-based and industrial Ethernet), the prediction system achieves real-time tracking of the status of each subsystem of the coke oven with millisecond-level accuracy, eliminating the delay caused by manual data entry. Simultaneously, the use of a Markush structure exhaustively limits all mainstream relational database types, ensuring broad software architecture compatibility and blocking technical circumvention at the underlying defense logic, thus ensuring persistent and secure data storage.

[0057] Please see Figure 4 In some embodiments, the present invention also provides a coking calorific value prediction system. This system is a physical functional module mapping of the above-mentioned prediction method to solve the problems of information silos between modules and the single function of simple post-event display in existing systems. The system includes: a data processing module 40, used to acquire information on the composition of coal entering the furnace, coal charging parameters, historical data on coking calorific value, and the set values ​​of preset parameters in the coking process parameters; a parameter feature value calculation module 41, used to calculate the amount of coal charged through the hole based on the coal charging parameters, and to calculate the calorific value of 1% hole impact based on the set values ​​of preset parameters; and an objective function value prediction module 42, used to input the information on the composition of coal entering the furnace, coal charging parameters, historical data on coking calorific value, the set values ​​of preset parameters, the amount of coal charged through the hole, and the calorific value of 1% hole impact into a pre-trained process mechanism model to predict the coking calorific value.

[0058] The architecture of this system adopts a design philosophy of high cohesion and low coupling. The data processing module 40 acts as a sensing organ, accurately aggregating the basic data stream and providing a static benchmark for prediction. The parameter feature value calculation module 41 acts as a primary computing engine to perform feature dimensionality reduction from macroscopic to microscopic sensitivity coefficients. Finally, the objective function value prediction module 42 solves the unified nonlinear mechanism formula, forming a complete closed loop of "sensing-benchmark-feature extraction-mechanism prediction", which enables complex prediction algorithms to be executed efficiently in industrial environments.

[0059] In some embodiments, the system further includes a trend tracking and comparison module, which automatically tracks and compares the fluctuation trends of the daily predicted and actual coking calorific value. The trend tracking and comparison module undertakes the management responsibility of "backward verification," mapping the predicted and actual calorific values ​​to trend charts through a built-in visualization engine. This helps technicians clearly determine the degree of agreement between the prediction model and the actual calorific value, thereby enabling reverse tracing of deviations and timely calibration of the baseline parameters in the data processing module, forming a virtuous cycle of self-optimization for the prediction system.

[0060] In some embodiments, the data processing module is specifically used to: automatically collect process parameter data during the coking production process through a programmable logic controller, a testing system, or an external system, and automatically save the collected data to a database.

[0061] In some embodiments, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the coking calorific value prediction methods described above. This device provides computational power in a physical form. The memory contains computer code capable of performing multi-dimensional calorific value compensation mechanism prediction. When the processor is activated, it can perform throughput of massive industrial data and nonlinear mapping calculations of the objective function, enabling the algorithm to be implemented in real hardware terminals such as edge computing gateways and industrial control computers.

[0062] In some embodiments, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the coking calorific value prediction methods described above. This storage medium encompasses physical carriers such as solid-state chips, hard disks, or optical discs capable of persistently or non-persistently storing algorithmic digital code, adapting to the independent circulation and distribution modes of modern industrial software.

[0063] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A coking heat consumption prediction method characterized by comprising: include: Obtain information on the composition of coal entering the furnace, coal charging parameters, historical data on coking calorific value, and the set values ​​of preset parameters in the coking process parameters; Calculate the coal loading amount based on the coal loading parameters. Calculate the heat loss due to 1% of the orifice based on the preset parameter settings; The information on the composition of the coal fed into the furnace, the information on the coal charging parameters, the historical data on coking calorific value, the set values ​​of the preset parameters, the amount of coal charged through the orifice, and the calorific value affected by 1% of the orifice are input into the pre-trained process mechanism model to predict the coking calorific value.

2. The coke consumption heat consumption prediction method according to claim 1, characterized by, The composition information of the coal fed into the furnace includes the moisture content and volatile matter content of the coal fed into the furnace. The coal loading parameter information includes the total coal loading amount and the number of furnaces loaded; the historical data of coking calorific value includes the lowest coking calorific value within a preset time period.

3. The coke consumption heat consumption prediction method according to claim 2, characterized by, The preset parameters in the coking process parameters include the standard value of coking calorific value, the standard coal loading amount of a single-hole carbonization chamber, the moisture content of standard wet coal, the volatile matter content of standard wet coal, the effect of 1% moisture content of coal entering the furnace on calorific value, and the effect of 1% volatile matter content of coal entering the furnace on calorific value.

4. The coke consumption heat consumption prediction method according to claim 2, characterized by, The step of calculating the coal loading amount per hole based on the coal loading parameter information includes: calculating the coal loading amount per hole based on the total coal loading amount and the number of furnaces loaded within a preset time period.

5. The coke consumption heat consumption prediction method according to claim 3, characterized by, The step of calculating the 1% hole-affected heat consumption based on the preset parameter settings includes: calculating the 1% hole-affected heat consumption based on the standard value of coking heat consumption and the standard coal loading amount of a single-hole carbonization chamber within a preset time period.

6. The coke consumption heat requirement prediction method according to claim 3, characterized by, The coal composition information, coal charging parameter information, historical coking calorific value data, preset parameter settings, orifice coal charging amount, and the calorific value influenced by 1% of the orifices are input into a pre-trained process mechanism model to predict the coking calorific value, including: Based on the aforementioned process mechanism model, the coking calorific value is calculated using the following parameters: coal charge per hole, moisture content of coal fed into the furnace, volatile matter content of coal fed into the furnace, moisture content of standard wet coal, volatile matter content of standard wet coal, standard coal charge per single-hole carbonization chamber, minimum coking calorific value within a preset time period, calorific value affected by 1% moisture content of coal fed into the furnace, calorific value affected by 1% volatile matter content of coal fed into the furnace, and calorific value affected by 1% of the hole.

7. The coke consumption heat requirement prediction method according to claim 1, characterized by, After predicting the coking calorific value, the method further includes: comparing the predicted coking calorific value with the target value of the coking calorific value to obtain the fluctuation monitoring result of the predicted coking calorific value; wherein the target value of the coking calorific value is determined by the average value of the predicted coking calorific value within a preset time period.

8. A coking heat consumption prediction system characterized by comprising: include: The data processing module is used to acquire information on the composition of coal fed into the furnace, coal charging parameters, historical data on coking calorific value, and the set values ​​of preset parameters in the coking process parameters. The parameter characteristic value calculation module is used to calculate the amount of coal loaded into the hole based on the coal loading parameter information, and to calculate the heat consumption affected by 1% of the hole based on the set value of the preset parameter. The objective function value prediction module is used to input the coal composition information, coal charging parameter information, historical data of coking calorific value, preset parameter settings, orifice coal charging amount, and the calorific value affected by 1% of the orifices into a pre-trained process mechanism model to predict the coking calorific value.

9. A computer device, comprising: The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the coking calorific value prediction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the coking heat consumption prediction method as described in any one of claims 1 to 7.

Images