A heat energy regulation system and method for co-producing pickled mechanism charcoal and activated carbon

Abstract

The present invention provides a heat energy regulation system and method for co-producing machine-made charcoal and activated carbon through pickling. It is applied to the technical field of comprehensive energy utilization. The system includes: a data acquisition module that collects heat source data during the production process of machine-made charcoal and heat demand data during the pickling process of activated carbon; a data preprocessing module for preprocessing the heat source data and the heat demand data to obtain two sub-sequences of heat supply and heat consumption; a heat regulation network model for respectively performing bidirectional modeling on the two sub-sequences of heat supply and heat consumption to extract heat flow characteristics, dynamically weighting the heat flow characteristics, thereby optimizing the energy transmission path and outputting an optimized dynamic heat distribution plan; a heat regulation module for, according to the optimized dynamic heat distribution plan, adjusting signals in real time to adjust the heat energy matching during the production of machine-made charcoal and the pickling process of activated carbon. The present invention effectively improves energy utilization efficiency, reduces energy consumption costs, and enhances comprehensive benefits.

Description

Technical Field

[0001] The present invention relates to the technical field of comprehensive energy utilization, and particularly to a heat energy regulation system and method for co-producing machine-made charcoal and activated carbon pickling. Background Art

[0002] With the increasing global attention to the efficient utilization of energy and environmental protection, the comprehensive energy utilization in all links of industrial production has become a research hotspot. In the production process of machine-made charcoal, a large amount of combustible gas rich in components such as methane, hydrogen, and carbon monoxide is generated by the high-temperature pyrolysis of biomass raw materials, and at the same time, a considerable amount of waste heat is released. In the pickling process of activated carbon, a large amount of steam is required for acid solution heating and activated carbon drying to ensure the pickling effect and the smooth progress of subsequent treatment processes. Theoretically, co-producing machine-made charcoal production and activated carbon pickling can achieve energy complementarity and efficient utilization, with significant economic and environmental benefits. However, in actual operation, there is a problem of asynchronous heat demand and supply in the production of machine-made charcoal and the pickling process, resulting in the waste of these precious energies and restricting the economic benefits of the co-production mode. Therefore, there is an urgent need for an intelligent energy allocation method to optimize heat energy matching, achieve efficient heat recovery and reuse, and improve energy utilization efficiency. Summary of the Invention

[0003] The present invention provides a heat energy regulation system and method for co-producing machine-made charcoal and activated carbon pickling, aiming to optimize the heat energy matching in the co-production process of machine-made charcoal production and activated carbon pickling through an intelligent heat regulation method, so as to solve the technical problems that traditional heat energy regulation methods are difficult to accurately match the heat demands of different links, resulting in heat asynchrony and energy waste.

[0004] To achieve the above object, it is realized through the following technical solutions:

[0005] According to the first aspect of the present invention, there is provided a heat energy regulation system for co-producing machine-made charcoal and activated carbon pickling, including:

[0006] A data acquisition module for real-time collecting heat source data in the production process of machine-made charcoal and heat demand data in the pickling process of activated carbon;

[0007] A data preprocessing module for normalizing and energy feature mapping the heat source data and the heat demand data, extracting energy-related features, and then performing data segmentation to obtain two subsequences of heat supply and heat consumption;

[0008] A heat regulation network model is used to perform bidirectional modeling on the two subsequences of heat supply and heat consumption respectively to extract heat flow characteristics, dynamically weight the heat flow characteristics corresponding to the heat energy distribution in different regions, ensure the efficient matching of the heat released by the mechanism carbon with the requirements of activated carbon pickling, thereby optimizing the energy transmission path, and output an optimized dynamic heat distribution plan;

[0009] A heat regulation module is used to adjust the heat energy matching in the process of mechanism carbon production and activated carbon pickling in real time according to the optimized dynamic heat distribution plan.

[0010] Further, among them, the heat regulation network model includes:

[0011] A tokenized Mamba encoder is used to perform bidirectional temporal modeling on the two subsequences of heat supply and heat consumption to extract basic heat flow characteristics;

[0012] A Gaussian decay mask dynamically assigns weights and optimizes heat energy matching according to the temperature gradient and conduction resistance in the heat energy transmission path to obtain weighted heat flow characteristics;

[0013] A semantic token learner is used to adaptively learn key heat flow characteristics;

[0014] A semantic token fuser fuses the encoded key heat flow characteristics with the original heat source data and outputs the fused efficient heat flow characteristics, and the efficient heat flow characteristics include global heat energy distribution and local optimization strategies;

[0015] A feature weighted fusion unit is used to fuse the basic heat flow characteristics and the efficient heat flow characteristics to obtain final optimized characteristics for realizing accurate modeling and optimization of heat energy transmission.

[0016] Further, among them, the tokenized Mamba encoder is used to perform bidirectional temporal modeling on the two subsequences of heat supply and heat consumption to extract basic heat flow characteristics, including:

[0017] Apply forward and backward convolution operations to these two subsequences respectively through bidirectional convolution processing, and then use one-dimensional convolution to extract features to obtain basic heat flow characteristics.

[0018] Further, among them, the Gaussian decay mask dynamically assigns weights and optimizes heat energy matching according to the temperature gradient and conduction resistance in the heat energy transmission path, including:

[0019] Calculate the spatial Gaussian distribution weight according to the index distance between each feature in the basic heat flow characteristics and the central node, and the standard deviation of the spatial Gaussian distribution;

[0020] Calculate the Gaussian distribution weights based on features according to the Euclidean distance between each feature in the basic heat flux features and the central node features, as well as the standard deviation of the Gaussian distribution based on features.

[0021] Multiply the spatial Gaussian distribution weights and the Gaussian distribution weights based on features to obtain the comprehensive weight of each feature.

[0022] Weight each original feature using the calculated comprehensive weights to obtain the weighted heat flux features.

[0023] Furthermore, the semantic token learner is used to adaptively learn key heat flux features, including:

[0024] Receive the heat flux features weighted by the Gaussian attenuation mask.

[0025] Use the cross-scanning method to merge two semi-directed sequences to obtain the integrated heat flux features, including: integrating two subsequences from different sources into a sequence of the original length through specific partitioning and splicing operations.

[0026] Compress the integrated heat flux feature data using the heat flux attention module, extract representative semantic tokens to obtain key heat flux features, and perform adaptive downsampling to optimize the thermal energy transfer efficiency.

[0027] Furthermore, the heat flux attention module includes a channel attention mechanism and a spatial attention mechanism, which are used to capture different heat flux patterns and spatial distribution features.

[0028] Learn key energy transfer patterns and generate semantic tokens through a series of heat flux pooling, feature convolution, and non-linear activation operations.

[0029] Perform adaptive downsampling on the generated semantic tokens, including screening or compressing the semantic tokens, to optimize the thermal energy transfer efficiency.

[0030] Process the downsampled semantic tokens using a multi-layer perceptron, and finally output the semantic tokens representing the heat flux pattern to obtain key heat flux features, where the key heat flux features include the thermal energy distribution in different regions, the change in heat flux density, and the heat conduction path information.

[0031] Furthermore, the semantic token fuser receives the semantic tokens representing the heat flux pattern and the original heat source data and performs feature fusion to form an overall thermal energy transfer sequence, and outputs the fused efficient heat flux features, making the fused efficient heat flux features more in line with the actual energy demand, while taking into account local details and global transfer efficiency, and realizing precise heat supply control for different process stages, including:

[0032]

[0033] Among them, is the fused high-efficiency heat flux feature; is the original input feature, that is, the original heat source data; is the weighted pooling operation to extract key features; is the semantic token learner operation; is the heat flux attention module operation; is the semantic token output by the semantic token learner.

[0034] Furthermore, among them, the heat flux attention module includes: a channel attention module and a thermal energy spatial attention module. The channel attention module and the thermal energy spatial attention module jointly act on the heat flux matching optimization process. By combining global and local information, the dynamic adjustment of the energy flow is realized, and the heat loss is reduced;

[0035] The feature weighted fusion unit inputs the fused features into a multi-layer perceptron for training the heat regulation network model, realizes the accurate modeling and optimization of the heat energy transmission, adjusts the local heat energy distribution in combination with the thermal energy spatial attention module, and uses the heat flux sensitive distance to optimize the loss function to improve the heat matching accuracy, and finally realizes the stability and efficiency of the heat supply control.

[0036] Furthermore, among them, the channel attention module extracts features along the channel dimension using global average pooling and max pooling, and generates a channel attention map through the processing of a multi-layer perceptron;

[0037] The thermal energy spatial attention module pools along the channel dimension to generate a feature map, splices and convolves to extract spatial attention features, and obtains a spatial attention map.

[0038] According to the second aspect of the present invention, a heat energy regulation method for co-producing machine-made charcoal and activated carbon pickling is also provided. This method includes:

[0039] Real-time collect the heat source data in the production process of machine-made charcoal and the heat demand data in the activated carbon pickling process;

[0040] Perform normalization and energy feature mapping processing on the heat source data and the heat demand data, extract the energy-related features, and then perform data segmentation to obtain two subsequences of heat supply and heat consumption;

[0041] Construct a heat regulation network model to perform bidirectional modeling on the two subsequences of heat supply and heat consumption respectively to extract heat flux features, dynamically weight the heat flux features corresponding to the heat energy distribution in different regions, ensure the efficient matching of the heat released by machine-made charcoal and the demand of activated carbon pickling, thereby optimizing the energy transmission path, and output an optimized dynamic heat distribution plan;

[0042] According to the optimized dynamic heat allocation scheme, real-time control signals are used to adjust the thermal energy matching in the production of mechanism charcoal and the pickling process of activated carbon.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] (1) The present invention proposes a tokenized Mamba encoder to perform bidirectional time series modeling on heat flow data. Starting from both ends of the heat source released by the cracking of mechanism charcoal and the heat demand of activated carbon pickling, a dynamic heat allocation mechanism is constructed. Through normalization processing and feature projection, key thermal energy features are extracted, and a one-dimensional convolutional network is used to process the heat supply and heat consumption subsequences. This method can accurately depict the spatio-temporal change trend of heat energy transmission, reduce energy waste, and improve the heat utilization efficiency.

[0045] (2) Traditional heat energy control methods are difficult to accurately match the heat demand of different links. The present invention proposes a Gaussian decay mask, which dynamically allocates the weights of each transmission step according to the temperature gradient and conduction resistance in the heat energy transmission path. By calculating the Gaussian decay weights of the spatial distribution and heat flow characteristics, the energy transmission path is optimized to ensure that the heat released by the mechanism charcoal can be efficiently transferred to the pickling link, thus achieving accurate heat matching.

[0046] (3) The present invention proposes a heat energy spatial attention module. By constructing a local heat energy distribution model, high-energy consumption areas and heat energy waste areas are accurately identified. The heat energy spatial attention module combines the channel attention mechanism and the spatial attention mechanism to extract features and perform weighted calculations on different heat flow areas, making the energy allocation more accurate, effectively reducing heat energy loss, and ensuring that the heat is concentrated in the key energy consumption areas, thereby improving the overall thermal efficiency.

[0047] (4) Traditional heat optimization methods are difficult to effectively measure the heat flow matching effect, resulting in large errors in heat allocation. The present invention proposes a heat flow sensitive distance as the optimization target, which combines the heat flow form weight and the scale adjustment factor to measure the heat flow matching degree between different links. The heat flow sensitive distance comprehensively considers the heat gradient, heat flow balance degree, and heat load change of the heat energy transmission path, and combines the aspect ratio error and the scale error to optimize the heat supply and demand balance and improve the heat energy utilization rate of the co-production process of mechanism charcoal and activated carbon pickling.

[0048] It should be understood that the content described in the summary of the invention is not intended to limit the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Brief Description of the Drawings

[0049] In conjunction with the accompanying drawings and with reference to the following detailed description, the above and other features, advantages, and aspects of the embodiments of the present invention will become more apparent. The accompanying drawings are used to better understand the solution and do not limit the present invention. In the drawings, the same or similar reference numerals denote the same or similar elements, where:

[0050] Figure 1 is a schematic diagram of the modules of a thermal energy regulation system for co-producing machine-made charcoal and activated carbon through pickling;

[0051] Figure 2 is a flowchart of a thermal energy regulation system for co-producing machine-made charcoal and activated carbon through pickling;

[0052] Figure 3 is a schematic diagram of the steps of a method for regulating thermal energy in the co-production of machine-made charcoal and activated carbon through pickling according to an embodiment of the present invention;

[0053] Figure 4 is a heat map of the visualization of the gradient distribution of the algorithm of the present invention;

[0054] Figure 5 is a graph showing the changing trends of the mean squared error, coefficient of determination, accuracy rate, and loss function of the algorithm of the present invention with the number of training rounds. Detailed Embodiments

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0056] In addition, the term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0057] Figure 1 shows a schematic diagram of the modules of a thermal energy regulation system for co-producing machine-made charcoal and activated carbon through pickling; Figure 2 shows a flowchart of a thermal energy regulation system for co-producing machine-made charcoal and activated carbon through pickling. As Figure 1 and Figure 2 shown, a system 100 for co-producing machine-made charcoal and activated carbon through pickling includes:

[0058] The data acquisition module 110 is used to collect heat source data during the production process of machine-made charcoal and heat demand data during the pickling process of activated carbon.

[0059] Among them, the heat source end data includes the time-series data of temperature, heat flux, and combustible gas components (such as methane, hydrogen, carbon monoxide, etc.) generated during the production process of machine-made charcoal. The heat demand end data includes the time-series demand data such as steam temperature, heat load, and drying time required for the activated carbon pickling process. During the specific acquisition process, a thermocouple (accuracy ±1°C) is used to collect the temperature of the machine-made charcoal cracking furnace (range 300 - 800°C) in real time, and a gas flow meter (measurement range 0 - 100 L / min) is used to monitor the flow rate of combustible gas ( 、 、CO), and a steam pressure sensor (0 - 10 bar) is used to collect the steam demand data in the pickling link.

[0060] The data preprocessing module 120 is used to perform normalization and energy feature mapping processing on the heat source data and the heat demand data, extract energy-related features, and then perform data segmentation to obtain two subsequences of heat supply and heat consumption. For example, the collected data is sampled at 5-second intervals, and Z-score normalization (mean ±3σ truncation) is performed.

[0061] The heat regulation network model 130 is used to perform bidirectional modeling on the two subsequences of heat supply and heat consumption respectively to extract heat flow features, dynamically weight the heat flow features corresponding to the heat energy distribution in different regions, ensure the efficient matching of the heat released by machine-made charcoal and the demand for activated carbon pickling, thereby optimizing the energy transmission path, and output an optimized dynamic heat distribution plan;

[0062] The heat regulation module 140 is used to, according to the optimized dynamic heat distribution plan, adjust the signal in real time to adjust the heat energy matching during the production process of machine-made charcoal and the pickling process of activated carbon.

[0063] For example, a 4 - 20 mA signal is output through a PID controller to adjust the opening of the waste heat boiler valve and the electric control valve of the steam distribution pipeline, and control the heat energy transmission rate.

[0064] The working process of the heat energy regulation system is as follows: 1. The data acquisition module obtains heat source and heat demand data in real time; 2. The data preprocessing module performs normalization and feature segmentation; 3. The heat regulation network model sequentially generates an optimization plan through encoding, weighting, token learning, and fusion; 4. The heat regulation module outputs a valve control signal. 5. Adjust the heat energy matching in real time through signals (such as valve opening, pump speed control instructions).

[0065] The co-production of machine-made charcoal and activated carbon by pickling uses water as a carrier to construct an energy storage and regulation system. With the high-temperature pyrolysis of machine-made charcoal, three heat sources are provided to meet the energy requirements for pickling. A heat regulation network model is used to predict and optimize the time intervals of energy supply and demand, effectively solving the problem of heat asynchrony, improving energy utilization efficiency, reducing energy consumption costs, achieving efficient collaborative production of machine-made charcoal and pickled activated carbon, and enhancing comprehensive benefits. This problem involves heat scheduling balance and is modeled based on heat supply and demand:

[0066]

[0067] : At time , the heat provided by machine-made charcoal production.

[0068] : At time , the heat required for the pickling process.

[0069] To achieve heat supply-demand matching in machine-made charcoal production and pickling, the present invention designs the following objective function for optimization. The optimization objective is to minimize the difference between heat supply and demand, that is:

[0070]

[0071] Among them, is the time period considered in the optimization process; represents the absolute value to ensure that the difference is positive.

[0072] Through the training of historical data for this optimization objective, the system can intelligently adjust the matching between energy supply and demand, thus eliminating the problem of heat asynchrony.

[0073] The data preprocessing module 120 is used to perform normalization and energy feature mapping processing (such as Fourier transform) on the heat source data and the heat demand data. After extracting the energy-related features, the data is segmented to obtain two subsequences of heat supply and heat consumption. Specifically, normalization uses Z-score standardization, and energy feature mapping extracts the frequency-domain energy distribution through Fourier transform, retaining the first 10 main frequency components to reduce the influence of noise.

[0074] This system adopts a centralized heat regulation method. Starting from the heat source released by the pyrolysis of mechanism carbon and the heat demand of activated carbon pickling, a dynamic heat allocation mechanism is constructed to achieve efficient utilization and synchronous regulation of energy. The system converts temperature and heat flow data into four groups of semi-directional time series features to more accurately capture the changing trends of heat supply and demand, reduce energy loss, and reduce the computational complexity through lightweight optimization, making it applicable to heat management under different working conditions. Specifically, the normalized heat source data and heat demand data are sliced in the time direction into: forward heat supply features (forward time series, heat source end), forward heat consumption features (forward time series, heat demand end), backward heat supply features (backward time series, heat source end), and backward heat consumption features (backward time series, heat demand end), namely four groups of semi-directional time series features.

[0075] By using two-way one-dimensional convolution to merge the forward and backward features, two subsequences of heat supply and heat consumption are split from the above four groups of semi-directional time series features, namely, a heat supply subsequence (forward + backward heat supply) and a heat consumption subsequence (forward + backward heat consumption) are generated.

[0076] Among them, the heat regulation network model 130 includes:

[0077] A tokenized Mamba encoder, which is used to perform two-way time modeling on the two subsequences of heat supply and heat consumption, and extract basic heat flow features;

[0078] A Gaussian decay mask, which dynamically assigns weights and optimizes heat energy matching according to the temperature gradient and conduction resistance in the heat energy transmission path, and obtains weighted heat flow features;

[0079] A semantic token learner, which is used to adaptively learn key heat flow features;

[0080] A semantic token fuser, which fuses the encoded key heat flow features with the original heat source data and outputs the fused efficient heat flow features. The efficient heat flow features include the global heat energy distribution and local optimization strategies;

[0081] A feature weighted fusion unit, which is used to fuse the basic heat flow features and the efficient heat flow features to obtain the final optimized features, and is used to achieve accurate modeling and optimization of heat energy transmission.

[0082] The tokenized Mamba encoder, Gaussian decay mask, semantic token learner, semantic token fuser, and feature weighted fusion unit included in the heat regulation network model gradually construct a precise dynamic heat allocation mechanism from the perspectives of feature extraction, heat energy modeling, semantic optimization, and feature fusion.

[0083] Preferably, in some embodiments, the tokenized Mamba encoder, which is used to perform two-way time modeling on the two subsequences of heat supply and heat consumption and extract basic heat flow features, includes:

[0084] The two subsequences are processed by bidirectional convolution, and forward and backward convolution operations are applied to them respectively. Then, one-dimensional convolution is used to extract features to obtain basic heat flow features.

[0085] Specifically, in terms of energy forecasting and regulation, a tokenized Mamba encoder is used to perform bidirectional modeling on the two subsequences of heat supply and heat consumption. The two subsequences of heat supply and heat consumption obtained by segmenting the time series heat flow data are respectively subjected to activation functions and one-dimensional convolution to extract features to obtain basic heat flow features, which are then input into the Gaussian attenuation mask of the control network. Subsequently, the Gaussian attenuation mask is used to optimize the thermal energy matching strategy, focusing on efficient energy transmission paths, and combined with the semantic token learner to adaptively learn key heat flow features for dynamic optimization and downsampling. Finally, the semantic token fuser is used to integrate multi-source energy information to achieve precise heating control, and the residual connection is used to improve system stability.

[0086] Among them, the key operation of the tokenized Mamba encoder is to encode the sequence through bidirectional processing, which is the basic sequence processing. It is based on the bidirectional temporal convolutional network (Bi-TCN), with forward and backward convolution kernel sizes of 3, step size 1, output channels of 64, and activation function ReLU. The formula is defined as: The formula includes:

[0087]

[0088] in, : Input feature sequence; : are the weight matrices for the forward and backward convolution operations, respectively, and are learnable parameters; : The corresponding bias term.

[0089] The tokenized Mamba encoder processes sequence data via bidirectional convolutions to capture contextual information in different directions.

[0090] The tokenized Mamba encoder is used to extract the time series features of the heat flow data, and the output is the normalized heat flow time series features, which are passed to the Gaussian attenuation mask to calculate the spatial weighted weights and optimize the thermal energy matching.

[0091] The tokenized Mamba encoder is responsible for extracting the temporal features of the heat flow data, while the Gaussian attenuation mask is mainly used for spatial feature modeling. The relationship between them can be understood as time-space joint modeling. The tokenized Mamba encoder generates the heat flow temporal features, which are used to calculate the heat demand of different regions at the current moment. The Gaussian attenuation mask uses these temporal features as input to dynamically adjust the distribution of thermal energy in space, optimize the transmission path, and improve the efficiency of heat utilization.

[0092] Preferably, in some embodiments, the Gaussian attenuation mask dynamically assigns weights and optimizes thermal energy matching according to the temperature gradient and conduction resistance in the thermal energy transfer path.

[0093] According to the characteristics of temperature gradient and conduction resistance in the heat transfer path, weights are assigned to each transfer step (referring to the discretization stage of heat energy in the heat supply path, that is, the propagation state of heat energy at different positions and time points), with a focus on key heat interaction nodes. Since heat flow has time dynamics (time series) and spatial diffusion characteristics (heat transfer path), it is necessary to optimize the modeling in both the time domain (tokenized Mamba encoder) and the spatial domain (Gaussian decay mask). By calculating the Gaussian decay weights of the spatial distribution and heat flow characteristics, a more refined heat energy scheduling is achieved, enhancing the system's perception and optimization ability of the energy transfer process. In the Gaussian decay mask mechanism, according to the temperature change rate and conduction efficiency in the heat transfer path, dynamic weighting is performed on the heat energy distribution in different regions to ensure that the heat released by the mechanism carbon is efficiently matched with the requirements of activated carbon pickling, thereby optimizing the energy transfer path and improving the overall heat utilization efficiency. Assume the feature sequence is where is the length of the feature sequence, that is, the total number of features. For the th feature , its weight is the product of the spatial Gaussian distribution weight and the feature-based Gaussian distribution weight:

[0094]

[0095] where is the spatial Gaussian distribution weight, and the calculation formula is:

[0096]

[0097] : the spatial Gaussian distribution weight of the th feature; : the index distance between the th feature and the central node; : the standard deviation of the spatial Gaussian distribution, controlling the speed of spatial decay, is set to 1 / 5 of the transmission path length to ensure that the weight covers the main heat conduction area.

[0098] where is the feature-based Gaussian distribution weight, and the calculation formula is:

[0099]

[0100] : the feature-based Gaussian distribution weight of the th feature; : the Euclidean distance between the th feature and the feature of the central node; : The standard deviation of the Gaussian distribution based on features controls the rate of feature decay.

[0101] Among them, the selection strategies for the central node include: (1) using the peak position of the heat demand as the central node to dynamically track the highest heat load area of the pickling process; (2) using the geometric center of the heat energy transmission path as the benchmark to ensure the balance of spatial weight distribution. Specifically, the system dynamically selects the central node according to the real-time heat load distribution: when there is a significant peak in the heat demand of the pickling process, the peak position is used as the central node; when the heat load distribution is uniform, the geometric center of the transmission path is used as the benchmark. Dynamic selection of the central node: Detect the local maximum value of the pickling heat demand sequence through a sliding window (window size: 10 time steps). If the peak exceeds the threshold (such as the mean + 2 times the standard deviation), the peak position is used as the center; otherwise, the geometric center of the transmission path (the midpoint of the sequence) is used as the benchmark.

[0102] Weight It is used for heat energy distribution during the pickling process of mechanism carbon and activated carbon. As the final feature weight, it dynamically weights the heat flow features, making the heat flow scheduling more accurate and improving the system's optimization ability for heat transfer.

[0103] Based on the temperature gradient and conduction resistance in the heat energy transmission path, the Gaussian decay mask calculates the weights of key heat flow interaction nodes to ensure the reasonable spatial distribution of heat flow features and optimize the perception ability of the heat energy regulation network. The output is the weighted weights of different feature points , which is transmitted to the semantic token learner to extract key heat flow features.

[0104] Preferably, in some embodiments, the semantic token learner is used to adaptively learn key heat flow features, including: receiving the heat flow features weighted by the Gaussian decay mask; using the cross-scanning method to merge two semi-directed sequences, such as using bidirectional LSTM or temporal convolution to process the semi-directed sequences to ensure that the temporal dynamics are not destroyed, and obtaining the integrated heat flow features, including: through specific partitioning and splicing operations, integrating two subsequences from different sources into a sequence of the original length, maintaining the temporal dynamics of the features, and providing a suitable input for the subsequent processing layer;

[0105] Using the heat flow attention module to compress the integrated heat flow feature data, extract representative semantic tokens, obtain key heat flow features, and perform adaptive downsampling to optimize the heat energy transmission efficiency.

[0106] Furthermore, the heat flow attention module compresses the heat flow feature data, extracts representative semantic tokens (key heat flow features), and performs adaptive downsampling to optimize the heat energy transfer efficiency. Through a series of heat flow pooling, feature convolution, and nonlinear activation operations, key energy transfer patterns are learned and semantic tokens are generated to capture efficient heat flow features and provide support for dynamic heat matching. The semantic token learner adaptively learns and extracts key heat flow features based on the heat flow attention module to achieve adaptive thermal optimization of different process sections. Assume that the input sequence is ,in, is the length of the input sequence, that is, the total number of input data or the number of samples, and the output token of the semantic token learner It can be expressed as:

[0107]

[0108] : The first elements; : Weighted features calculated by the heat flow attention module; MLP: Multi-layer perceptron, used to process the features output by the attention mechanism and generate the final semantic token.

[0109] Through cross-scanning and feature concatenation, the semantic token learner extracts representative semantic tokens and obtains key heat flow features to achieve efficient heat matching. The output is a semantic token (representing the heat flow pattern), which is passed to the semantic token fuser for feature fusion.

[0110] Preferably, in some embodiments, the heat flow attention module includes a channel attention mechanism and a spatial attention mechanism for capturing different heat flow patterns and spatial distribution characteristics;

[0111] Through a series of heat flow pooling, feature convolution and nonlinear activation operations, key energy transfer patterns are learned and semantic tokens are generated;

[0112] Adaptively downsampling the generated semantic tokens, including filtering or compressing the semantic tokens to optimize thermal energy transfer efficiency;

[0113] The downsampled semantic tokens are processed using a multi-layer perceptron, and finally semantic tokens representing the heat flow pattern are output to obtain key heat flow features, which include thermal energy distribution in different regions, changes in heat flux density, and heat conduction path information.

[0114] Preferably, in some embodiments, the semantic token fuser receives semantic tokens representing the heat flow pattern and the original heat source data, performs feature fusion to form an overall thermal energy transfer sequence, and outputs the fused efficient heat flow features, making the fused efficient heat flow features more in line with the actual energy requirements, while taking into account local details and global transfer efficiency, and achieving precise heating control for different process stages, including:

[0115] During the heating control process, the encoded semantic heat flow features are fused with the original heat source data such as temperature, flow rate and other information. Through operations such as adaptive pooling, weight assignment calculation and weighted summation, the semantic token information is incorporated into the overall thermal energy transfer sequence, making the fused features more in line with the actual energy requirements, while taking into account local details and global transfer efficiency. The semantic token fuser finally integrates multi-source energy information, optimizes the heat flow matching, realizes precise heating control, and improves the system stability through residual connection, enhancing the adaptability to heat load fluctuations.

[0116] The formula for its fusion process includes:

[0117]

[0118] Wherein, is the fused efficient heat flow feature; is the original input feature, that is, the original heat source data; is the weighted pooling operation to extract key features; is the semantic token learner operation; is the heat flow attention module operation; is the semantic token output by the semantic token learner.

[0119] Combining the semantic heat flow features with the original heat source data such as temperature, flow rate and other information, the semantic token fuser integrates multi-source heat flow information for global optimization and matching to ensure that the heat flow transmission meets the actual requirements. The output is the fused efficient heat flow feature, which is transmitted to the feature weighted fusion unit for final training and optimization.

[0120] In the thermal energy regulation task, heat transfer is affected by both channel characteristics (different heat flux patterns) and spatial characteristics (the distribution of heat flux in different regions). Therefore, the heat flux attention module consists of a channel attention module and a thermal energy spatial attention module. The two work together in the heat flux matching optimization process to improve the heat matching efficiency in a global-local joint optimization manner, reduce heat loss, and achieve dynamic adjustment of the energy flow. Among them, the channel attention module extracts key heat flux characteristics along the direction of the heat conduction path using global average pooling and max pooling, and calculates and generates heat flux distribution weights through a multi-layer perceptron to optimize the thermal energy utilization efficiency of different technological processes. The thermal energy spatial attention module accurately captures local hot spots and inefficient regions in heat transfer through local energy distribution modeling, realizes dynamic adjustment of the energy flow, and reduces heat loss.

[0121] The channel attention module extracts features along the channel dimension using global average pooling and max pooling, and generates a channel attention map through the processing of a multi-layer perceptron, including:

[0122] Perform global average pooling and global max pooling Combine them into , and generate a channel attention map after being processed by a two-layer multi-layer perceptron :

[0123]

[0124] Among them, is the channel attention map, with a dimension of , serving as channel-level weight assignment, adjusting the heat flux contribution of each channel, evaluating the global importance of different heat flux characteristics, and optimizing the thermal energy utilization efficiency of different technological processes; is the activation function; MLP is a multi-layer perceptron, including a dimensionality reduction layer and a dimensionality increase layer.

[0125] Input feature , is the number of channels, is the height of the feature map, is the width of the feature map, obtained through global average pooling (GAP) and max pooling (GMP) . The MLP structure is two-layer fully connected. The first layer reduces the C dimension to C / 4 and activates with ReLU; the second layer restores to the C dimension and activates with Sigmoid to generate the channel attention map .

[0126] The thermal energy spatial attention module generates a feature map through channel - dimension pooling, concatenates it, and extracts spatial attention features through convolution. The thermal energy spatial attention module transforms it into a bidirectional two - dimensional spatial attenuation format. Through the spatial attenuation matrix, the greater the attenuation degree of the attention score of the surrounding markers farther away from the target center, so as to focus more on the target center and its surrounding areas while paying attention to global information.

[0127] Specifically, the thermal energy spatial attention module generates a feature map through channel - dimension pooling, concatenates it, and extracts spatial attention features through convolution, that is, average pooling along the channel dimension and max pooling along the channel dimension are concatenated into , and then pass through a convolutional layer to obtain the spatial attention map :

[0128]

[0129] Among them, is the spatial attention map, with a dimension of , serving as the attention weight at the spatial level, optimizing the matching effect of the thermal energy flow in different regions, optimizing local thermal energy transmission, paying attention to thermal energy hotspots and inefficient regions, and improving the thermal energy utilization rate; is the feature map obtained by concatenating the average pooling along the channel dimension and the max pooling along the channel dimension ; is the activation function, usually the Sigmoid function; Conv is the convolution operation.

[0130] Assume the input feature map is , the average pooling along the channel dimension and the max pooling along the channel dimension are respectively:

[0131]

[0132] Among them, is the feature map after average pooling along the channel dimension, with a dimension of , used to extract the spatial thermal energy flow distribution features; is the cumulative operation on all elements in the rows and columns of the feature map; is the normalization factor, calculating the pixel mean value, making the output independent of the size of the input feature map; represents the th channel, the th row, and the th column element value in the input feature map. Among them, ranges from 1 to ( is the number of channels), ranges from 1 to ( is the height of the feature map), The value range is 1 to ( is the width of the feature map); is the feature map after channel maximum pooling, the dimension is ; Used to extract spatial heat flow distribution characteristics; Represents a dual maximum operation, finding the maximum value in the feature structure.

[0133] Average pooling of channels and channel max pooling Splice to , ; 7×7 convolution kernel processing, output single channel feature map, Sigmoid activation to generate spatial attention map .

[0134] The feature weighted fusion unit inputs the fused features into the multi-layer perceptron to train the heat control network model, realizes the accurate modeling and optimization of heat energy transmission, adjusts the local heat energy distribution in combination with the thermal energy space attention module, and uses the heat flow sensitive distance to optimize the loss function, so as to improve the heat matching accuracy and finally realize the stability and efficiency of heating control.

[0135] In order to optimize the heat energy matching in the process of co-production of machine-made charcoal and activated carbon pickling, this step achieves accurate modeling and optimization of heat energy transfer by fusing the original heat flow features with the efficient heat flow features extracted by the semantic token fuser. The fused features are input into the multi-layer perceptron to train the heat control network model to ensure that the heat released by the machine-made charcoal can be efficiently transferred to the activated carbon pickling process. In addition, the local heat energy distribution is adjusted in combination with the thermal energy space attention module, and the heat flow sensitive distance is used to optimize the loss function to improve the heat matching accuracy, ultimately achieving the stability and efficiency of heating control.

[0136] Among them, the heat matching loss function is optimized:

[0137] The heat flow morphology weight and dynamic scale adjustment factor are introduced to calculate the heat flow sensitive distance based on the heat gradient of the heat energy transmission path, the balance of heat flow distribution and the heat load changes at different stages, and the heat matching error term is included to enhance the system's adaptability to heat imbalance problems and improve the thermal energy utilization rate of the co-production process of machine-made carbon and activated carbon acid washing. In order to optimize the accuracy of the target, the loss function includes the following parts:

[0138]

[0139] in, is the heat matching loss function, is the weight factor of the shape heat flux sensitive distance, used to adjust the influence of the loss; is the heat flux sensitive distance.

[0140] The calculation formula for defining the heat flux sensitive distance is:

[0141]

[0142] Among them, is the generalized intersection over union loss; is the heat flux sensitive distance; is the heat flux morphology weight, controlling the importance of the aspect ratio error; is the scale adjustment factor, controlling the importance of the scale error; is the aspect ratio error; in some specific embodiments, for example, it is set to: , β = 0.5.

[0143]

[0144] Among them, P is the smallest bounding box of B and The IoU is the intersection over union.

[0145] Let the heat supply and demand time window be , and the heat supply and demand reference window be , where is the center coordinate of the box, is the width and height of the box. Define the aspect ratio error The calculation formula is:

[0146]

[0147] represents the aspect ratio error, that is, the scale error, measuring the difference in shape between the heat supply and demand time window and the heat supply and demand reference window, especially the change in the aspect ratio. represents calculating the aspect ratio of the heat supply and demand reference window, taking the larger value to ensure unified scale measurement of the aspect ratio change. represents calculating the aspect ratio of the heat supply and demand time window and taking the maximum value among them to ensure numerical stability. is a very small constant (such as 10-7), preventing division by zero errors.

[0148] Scale error The calculation formula is:

[0149]

[0150] Among them, is the scale error, measuring the proportional difference between the size of the heat supply and demand time window and the size of the heat supply and demand reference window; is the maximum size (either width or height) of the heat supply and demand benchmark window; is the maximum size (either width or height) of the heat supply and demand time window.

[0151] In summary, the tokenized Mamba encoder extracts temporal heat flow features to form a preliminary input. The Gaussian decay mask calculates spatial weights to optimize the heat energy matching strategy. The semantic token learner extracts key heat flow features, generates semantic tokens, and enhances the learning of energy transmission patterns. The semantic token fusion unit combines the original input, optimizes the global heat energy regulation strategy, and improves the accuracy of heat supply. The feature weighted fusion unit performs the final optimization and trains the MLP to achieve stable control of the heat regulation network model. Finally, these five parts work together to ensure the optimal heat energy matching of the mechanism charcoal and activated carbon pickling process and improve the overall heat utilization rate.

[0152] After the above processing, the algorithm finally outputs an optimized heat supply and demand matching scheme to achieve efficient heat energy utilization in the production of mechanism charcoal and the pickling process of activated carbon. The algorithm performs dynamic energy management through the heat regulation network model 130, optimizes heat flow matching in combination with the Gaussian decay mask, and uses the semantic token learner and semantic token fusion unit to integrate multi-source energy information, optimize the heat transfer path, and ensure that heat energy flows along the efficient energy transfer path through Gaussian decay mask optimization, reduce heat loss, and improve the overall heat energy utilization rate. Extract key heat flow features through the semantic token learner and perform multi-layer energy fusion in combination with the semantic token fusion unit to achieve precise heat supply regulation in different process stages and reduce heat waste. The system finally minimizes the heat matching loss function between heat supply and heat consumption, improves energy utilization efficiency, and reduces operating costs.

[0153] The above embodiments of the present invention perform two-way temporal modeling through the tokenized Mamba encoder, accurately depict the spatio-temporal change trend of heat energy transmission, and reduce energy waste. The Gaussian decay mask dynamically assigns weights, optimizes the energy transmission path, and achieves precise heat matching. The heat energy spatial attention module combines the channel attention mechanism and the spatial attention mechanism to accurately identify high-energy consumption areas and reduce heat energy loss. The heat flow sensitive distance is used as the optimization target to improve the accuracy of heat supply and demand matching and enhance the overall heat efficiency.

[0154] The above is the introduction of the system embodiment. The following further illustrates the solution of the present invention through method embodiments.

[0155] Figure 3 shows a schematic diagram of the steps of a method for co-producing mechanism charcoal and pickling activated carbon according to an embodiment of the present invention. As Figure 3 shown, a method for co-producing mechanism charcoal and pickling activated carbon includes:

[0156] S201: Collect the heat source data during the production process of mechanism charcoal and the heat demand data during the pickling process of activated carbon in real time;

[0157] S202: Perform normalization and energy feature mapping processing on the heat source data and the heat demand data, extract the energy-related features, and then perform data segmentation to obtain two subsequences of heat supply and heat consumption;

[0158] S203: Construct a heat regulation network model to perform bidirectional modeling on the two subsequences of heat supply and heat consumption respectively to extract heat flow features, dynamically weight the heat flow features corresponding to the heat energy distribution in different regions, ensure the efficient matching of the heat released by mechanism charcoal and the demand for pickling activated carbon, thereby optimizing the energy transmission path, and output an optimized dynamic heat distribution plan;

[0159] S204: According to the optimized dynamic heat distribution plan, adjust the real-time control signal to adjust the heat energy matching during the production of mechanism charcoal and the pickling process of activated carbon.

[0160] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the described steps can refer to the corresponding processes in the foregoing system embodiments, and will not be elaborated herein.

[0161] The experimental results are as follows:

[0162] As Figure 4 shown in the heat map of the visualized gradient distribution, which shows the gradient changes in different regions, that is, the heat map of the Gaussian decay mask weight distribution, and can show that the heat energy is preferentially allocated to high-demand regions to evaluate the effect of feature learning. The color represents the intensity of the heat flow gradient, the red area represents a higher gradient change, and the blue area represents a lower gradient change. This visualization helps to analyze the key regions in the process of optimizing heat energy transmission.

[0163] The gradient mean of the overall gradient distribution is 0.045, and the gradient standard deviation is 0.012. The maximum gradient value is approximately 0.092; the minimum gradient value is approximately 0.008. The proportion of high-gradient regions as key points of energy transmission (value > 0.8) is 13.2%, indicating that the heat energy change in this region is the most intense. The proportion of low-gradient regions as heat stable regions (value < 0.02) is 27.5%, indicating that the heat in this region is relatively uniform and the optimization space is limited.

[0164] Under the action of the Gaussian decay mask optimization strategy, the heat flux gradient shows a trend of gradually decreasing from the high-gradient center outward, which conforms to the theoretical expectation of the gradual and balanced heat transfer. The semantic token learner extracts the main key heat flux features, indicating that the dynamic optimization has a significant effect in these regions. During the process of integrating energy information, the semantic token fuser reduces the overall gradient variance by about 23.7%, indicating that the heat supply control is more stable.

[0165] Figure 5 Shows the changing trends of the mean square error, coefficient of determination, accuracy, and loss function of the algorithm with the number of training rounds. Among them, the mean square error uses yellow dots, the coefficient of determination uses orange squares, the accuracy uses red triangles, and the heat matching loss function uses pink diamonds. Among them, for network training: Adam optimizer (learning rate 0.001, decay rate ×0.1 every 50 rounds), batch size 32, training for 200 rounds, and the loss function is . The training curve shows that the loss function converges, and the accuracy of the validation set reaches 92%.

[0166] The convergence speed of the model is relatively fast and enters a stable state around 30 rounds. Continuing training has limited benefits. The period between 10 - 25 rounds is the key learning stage, where the mean square error drops the fastest, and the accuracy and coefficient of determination increase the most significantly, indicating that the learning effect in this stage is the most remarkable. The final accuracy reaches 98%, the coefficient of determination is close to 0.9, and the heat matching loss function is below 0.05, and the mean square error is below 0.1, indicating that the model already has high prediction accuracy and stability. For practical applications, it is recommended to train for about 30 rounds because at this time the model performance has tended to be stable, and increasing the number of training rounds has little improvement on the results, which can save computing resources.

[0167] According to Figure 5 the experimental results show that the algorithm tends to be stable after 30 rounds of training, improves the energy utilization rate, and realizes the efficient collaborative production of mechanism charcoal and activated carbon pickling.

[0168] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in the present invention can be achieved. No limitations are imposed herein.

[0169] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A heat energy regulation system for co-producing acid-washed mechanism charcoal and activated carbon, characterized in that, It includes: A data acquisition module for real-time acquisition of heat source data in the process of mechanism charcoal production and heat demand data in the process of activated carbon pickling; A data preprocessing module for normalizing and energy feature mapping processing of the heat source data and the heat demand data, extracting energy-related features and then performing data segmentation to obtain two subsequences of heat supply and heat consumption; A heat regulation network model for bidirectional modeling of the two subsequences of heat supply and heat consumption respectively to extract heat flow features, dynamically weighting the heat flow features corresponding to the heat energy distribution in different regions to ensure efficient matching of the heat released by mechanism charcoal and the demand of activated carbon pickling, thereby optimizing the energy transmission path and outputting an optimized dynamic heat distribution plan; Among them, the heat regulation network model includes: A tokenized Mamba encoder for bidirectional time modeling of the two subsequences of heat supply and heat consumption to extract basic heat flow features; A Gaussian decay mask for dynamically allocating weights and optimizing heat energy matching according to the temperature gradient and conduction resistance in the heat energy transmission path to obtain weighted heat flow features; A semantic token learner for adaptively learning key heat flow features; A semantic token fuser for fusing the encoded key heat flow features with the original heat source data and outputting fused efficient heat flow features, where the efficient heat flow features include global heat energy distribution and local optimization strategies; A feature weighted fusion unit for fusing the basic heat flow features and the efficient heat flow features to obtain final optimized features for realizing accurate modeling and optimization of heat energy transmission; A heat regulation module for real-time regulating signals according to the optimized dynamic heat distribution plan to adjust the heat energy matching in the process of mechanism charcoal production and activated carbon pickling.

2. The heat energy regulation system for co-producing acid-washed mechanism charcoal and activated carbon according to claim 1, characterized in that Among them, The tokenized Mamba encoder for bidirectional time modeling of the two subsequences of heat supply and heat consumption to extract basic heat flow features includes: Applying forward and backward convolution operations to these two subsequences respectively through bidirectional convolution processing, and then using one-dimensional convolution to extract features to obtain basic heat flow features.

3. The heat energy regulation system for co-producing pickled mechanism charcoal and activated carbon according to claim 1, wherein Among them, The Gaussian decay mask for dynamically allocating weights and optimizing heat energy matching according to the temperature gradient and conduction resistance in the heat energy transmission path includes: Calculating spatial Gaussian distribution weights according to the index distance between each feature in the basic heat flow features and the central node, and the standard deviation of the spatial Gaussian distribution; Calculating Gaussian distribution weights based on features according to the Euclidean distance between each feature in the basic heat flow features and the central node feature, and the standard deviation of the feature-based Gaussian distribution; Multiplying the spatial Gaussian distribution weights and the feature-based Gaussian distribution weights to obtain the comprehensive weight of each feature; Using the calculated comprehensive weights to weight each original feature to obtain weighted heat flow features.

4. The heat energy regulation system for co-producing acid-washed mechanism charcoal and activated carbon according to claim 1, wherein Among them, The semantic token learner for adaptively learning key heat flow features includes: Receiving the heat flow features weighted by the Gaussian decay mask; Using the method of cross-scanning to merge two semi-directed sequences to obtain integrated heat flow features, including: integrating two subsequences from different sources into a sequence of the original length through specific partitioning and splicing operations; Compress the integrated heat flow feature data using a heat flow attention module, extract representative semantic tokens to obtain key heat flow features, and perform adaptive downsampling to optimize the heat energy transmission efficiency.

5. The heat energy regulation system for co-producing acid-washed mechanism charcoal and activated carbon according to claim 4, characterized in that, Among them, the heat flow attention module includes a channel attention mechanism and a spatial attention mechanism, which are used to capture different heat flow patterns and spatial distribution features; Through a series of heat flow pooling, feature convolution, and non-linear activation operations, learn the key energy transmission patterns and generate semantic tokens; Perform adaptive downsampling on the generated semantic tokens, including screening or compressing the semantic tokens, to optimize the heat energy transmission efficiency; Use a multi-layer perceptron to process the downsampled semantic tokens, and finally output the semantic tokens representing the heat flow pattern to obtain key heat flow features. The key heat flow features include the heat energy distribution in different regions, the change in heat flow density, and the heat conduction path information.

6. The heat energy regulation system for co-producing acid-washed mechanism charcoal and activated carbon according to claim 1, wherein, Among them, the semantic token fuser receives the semantic tokens representing the heat flow pattern and the original heat source data and performs feature fusion to form an overall heat energy transmission sequence, and outputs the fused efficient heat flow features, including: ， Among them, is the fused high-efficiency heat flux feature; is the original input feature, that is, the original heat source data; is the weighted pooling operation to extract key features; STL is the semantic token learner operation; is the heat flux attention module operation; is the semantic token output by the semantic token learner.

7. The heat energy regulation system for co-producing acid-washed mechanism charcoal and activated carbon according to claim 6, wherein Among them, the heat flow attention module includes: a channel attention module and a heat energy spatial attention module. The channel attention module and the heat energy spatial attention module jointly act on the heat flow matching optimization process. By combining global and local information, the dynamic adjustment of the energy flow is realized, and the heat loss is reduced; The feature weighted fusion unit inputs the fused features into a multi-layer perceptron for training the heat regulation network model, realizes the precise modeling and optimization of the heat energy transmission, adjusts the local heat energy distribution in combination with the heat energy spatial attention module, and uses the heat flow sensitive distance to optimize the loss function to improve the heat matching accuracy, and finally realizes the stability and efficiency of the heat supply control.

8. The heat energy regulation system for co-producing acid-washed mechanism charcoal and activated carbon according to claim 7, wherein Among them, the channel attention module extracts features along the channel dimension using global average pooling and max pooling, and generates a channel attention map through the processing of a multi-layer perceptron; The heat energy spatial attention module pools along the channel dimension to generate a feature map, splices and convolves to extract spatial attention features, and obtains a spatial attention map.

9. A heat energy regulation method for co-producing acid-washed mechanism carbon and activated carbon, characterized in that This method includes: Real-time collection of heat source data in the process of mechanism charcoal production and heat demand data in the process of activated carbon pickling; Perform normalization and energy feature mapping processing on the heat source data and the heat demand data, extract energy-related features and then perform data segmentation to obtain two subsequences of heat supply and heat consumption; Construct a heat regulation network model, which is used to perform bidirectional modeling on the two subsequences of heat supply and heat consumption respectively to extract heat flow features, dynamically weight the heat flow features corresponding to the heat energy distribution in different regions, ensure the efficient matching of the heat released by the mechanism charcoal and the demand of the activated carbon pickling, so as to optimize the energy transmission path, and output an optimized dynamic heat distribution plan; Among them, the heat regulation network model includes: A tokenized mamba encoder, which is used to perform bidirectional temporal modeling on the two subsequences of heat supply and heat consumption to extract basic heat flow features; A Gaussian decay mask, which dynamically assigns weights and optimizes heat energy matching according to the temperature gradient and conduction resistance in the heat energy transmission path to obtain weighted heat flow features; A semantic token learner for adaptively learning key heat flow features; A semantic token fusion unit that fuses the encoded key heat flow features with the original heat source data and outputs the fused efficient heat flow features, where the efficient heat flow features include global thermal energy distribution and local optimization strategies; A feature weighted fusion unit for fusing the basic heat flow features and the efficient heat flow features to obtain the final optimized features for realizing precise modeling and optimization of heat energy transmission; According to the optimized dynamic heat distribution scheme, the signal is adjusted in real time to adjust the heat energy matching during the production of mechanism charcoal and the pickling process of activated carbon.

Images