Carbon emission monitoring method and device applied to cement production line, computer equipment, readable storage medium and program product

By constructing a joint feature vector in a cement production line and utilizing the Adam-DBN model, the accuracy and robustness issues of traditional carbon emission monitoring methods are solved, achieving efficient and accurate carbon emission monitoring and equipment management.

CN118966548APending Publication Date: 2024-11-15南方电网能源发展研究院有限责任公司 +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411099962.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Traditional carbon emission monitoring methods in cement production lines suffer from low measurement accuracy, high cost, susceptibility to external interference, and difficult equipment maintenance, especially in high-temperature and complex emission gas environments where accurate monitoring is difficult to achieve.

Method used

By acquiring non-electrical characteristic data, equipment status data, and equipment power consumption data of key equipment in cement production lines, a joint feature vector is constructed. Using load monitoring models and carbon emission monitoring models, combined with the Adam-DBN model for training and optimization, the monitoring of changes in power load and material characteristics is realized, and the total carbon emissions are finally calculated.

Benefits of technology

It improves the accuracy of carbon emission measurement and the efficiency of equipment management, solves the accuracy and robustness problems of traditional methods, and realizes efficient and accurate carbon emission monitoring and equipment control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118966548A_ABST
    Figure CN118966548A_ABST
Patent Text Reader

Abstract

The invention relates to a carbon emission monitoring method, device and equipment applied to a cement production line, a storage medium and a program product, and relates to the technical field of carbon emission monitoring. According to the invention, the measurement accuracy of the carbon emission can be improved. The method comprises the following steps: acquiring non-electrical characteristic data, equipment state data and equipment power consumption data of key preset type equipment in a cement production line, and integrating and sorting the non-electrical characteristic data, the equipment state data and the equipment power consumption data to obtain a target data set; according to the target data set, constructing a joint feature vector corresponding to each piece of non-electrical feature data; inputting the joint feature vector into a load monitoring model to obtain power load information of the equipment; equipment operation information of the key preset type equipment and material characteristic change information in the production process of the cement production line are obtained, the power load information, the equipment operation information and the material characteristic change information are input into the carbon emission monitoring model, and the sum of carbon emission generated by the cement production line is obtained.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of carbon emission monitoring, and in particular to a carbon emission monitoring method, device, computer equipment, computer-readable storage medium and computer program product applied to a cement production line. Background Art

[0002] Cement production lines are key industrial production systems used to produce cement and related building materials. However, in the face of the current global trend of paying attention to carbon emissions, cement manufacturers and other related companies need to effectively monitor and manage their carbon emissions.

[0003] Among traditional technologies, the carbon emission monitoring technologies commonly used by cement production enterprises include direct measurement and indirect measurement. The direct measurement method usually involves high initial investment, operating costs and labor costs due to the installation and maintenance of online monitoring equipment such as flue gas analyzers, and external interference from factors such as dust, humidity and other gas components in the flue gas will also affect the accuracy of the measurement results; the indirect measurement method relies on accurate recording of raw material consumption and energy consumption, but there may be recording errors in actual operations, especially in production links with high material mobility. Energy consumption calculations usually require the use of specific carbon emission factors for conversion, but these factors may not accurately reflect the actual situation, resulting in the problem of low measurement accuracy in both of the above two measurement methods. Summary of the invention

[0004] Based on this, it is necessary to provide a carbon emission monitoring method, device, computer equipment, computer readable storage medium and computer program product applied to a cement production line to address the above technical problems.

[0005] In a first aspect, the present application provides a carbon emission monitoring method applied to a cement production line, comprising:

[0006] Acquire non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical characteristic data, the equipment status data and the equipment power consumption data to obtain a target data set;

[0007] According to the target data set, constructing joint feature vectors corresponding to each of the non-electrical feature data;

[0008] Inputting the joint feature vector into the trained load monitoring model to obtain the power load information of the key preset type of equipment;

[0009] The equipment operation information of the key preset types of equipment and the material characteristic change information during the production process of the cement production line are obtained, and the power load information, the equipment operation information and the material characteristic change information are input into a carbon emission monitoring model to obtain the total carbon emissions generated by the cement production line.

[0010] In one embodiment, the non-electrical characteristic data, the device status data, and the device power consumption data are integrated and sorted to obtain a target data set, including:

[0011] The non-electrical characteristic data, the device status data and the device power consumption data are integrated at the same time interval; the integrated data are arranged in chronological order to form a current data set including a timestamp column and a corresponding indicator column; outliers are removed and missing values ​​are filled in the current data set to obtain the target data set.

[0012] In one embodiment, before inputting the joint feature vector into the trained load monitoring model, the method further includes:

[0013] The target data set is divided into a training set and a test set according to a preset ratio; the training set is input into the load monitoring model to be trained for iterative training to obtain a preliminarily trained load monitoring model; the preliminarily trained load monitoring model is tested and tuned using the test set until the preliminarily trained load monitoring model meets the accuracy and stability conditions, thereby obtaining the trained load monitoring model.

[0014] In one embodiment, the step of inputting the training set into the load monitoring model to be trained for iterative training to obtain a preliminarily trained load monitoring model includes:

[0015] The first neural network model is trained using the training set, and the weight and offset of the first neural network model are determined; based on the weight and the offset, the state information of the hidden neurons of the first neural network model is obtained, the state information is used as the input vector of the second neural network model, and the second neural network model is tuned according to the input vector; the first neural network model and the second neural network model are adjusted using a preset optimizer to obtain the preliminarily trained load monitoring model.

[0016] In one embodiment, before inputting the power load information, the equipment operation information and the material characteristic change information into the carbon emission monitoring model, the method further includes:

[0017] Construct an initial carbon emission monitoring model including an input layer, a merging layer, a fully connected layer and an output layer; determine a target mapping relationship between carbon emissions and the power load information, the equipment operation information and the material characteristic change information based on the power load information, the equipment operation information and the material characteristic change information; according to the target mapping relationship, adjust the model parameters of the initial carbon emission monitoring model to obtain the carbon emission monitoring model.

[0018] In one embodiment, the method further comprises:

[0019] Obtain evaluation values ​​of the carbon emission monitoring model under multiple preset evaluation indicators, and determine whether the evaluation values ​​meet the corresponding threshold conditions; when any of the evaluation values ​​does not meet the threshold conditions, determine the target model parameters corresponding to the current evaluation indicator, and adjust the target model parameters until the evaluation value of the current evaluation indicator meets the threshold conditions.

[0020] In a second aspect, the present application also provides a carbon emission monitoring device applied to a cement production line, comprising:

[0021] A data processing module, used to obtain non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical characteristic data, the equipment status data and the equipment power consumption data to obtain a target data set;

[0022] A vector construction module, used for constructing joint feature vectors corresponding to each of the non-electrical feature data according to the target data set;

[0023] An information acquisition module, used for inputting the joint feature vector into the trained load monitoring model to obtain the power load information of the key preset type of equipment;

[0024] The carbon emission monitoring module is used to obtain the equipment operation information of the key preset types of equipment and the material characteristic change information during the production process of the cement production line, input the power load information, the equipment operation information and the material characteristic change information into the carbon emission monitoring model, and obtain the total carbon emissions generated by the cement production line.

[0025] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0026] Obtain non-electrical feature data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical feature data, the equipment status data and the equipment power consumption data to obtain a target data set; construct a joint feature vector corresponding to each of the non-electrical feature data according to the target data set; input the joint feature vector into a trained load monitoring model to obtain power load information of the key preset types of equipment; obtain equipment operation information of the key preset types of equipment and material feature change information in the production process of the cement production line, input the power load information, the equipment operation information and the material feature change information into a carbon emission monitoring model to obtain the total carbon emissions generated by the cement production line.

[0027] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0028] Obtain non-electrical feature data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical feature data, the equipment status data and the equipment power consumption data to obtain a target data set; construct a joint feature vector corresponding to each of the non-electrical feature data according to the target data set; input the joint feature vector into a trained load monitoring model to obtain power load information of the key preset types of equipment; obtain equipment operation information of the key preset types of equipment and material feature change information in the production process of the cement production line, input the power load information, the equipment operation information and the material feature change information into a carbon emission monitoring model to obtain the total carbon emissions generated by the cement production line.

[0029] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0030] Obtain non-electrical feature data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical feature data, the equipment status data and the equipment power consumption data to obtain a target data set; construct a joint feature vector corresponding to each of the non-electrical feature data according to the target data set; input the joint feature vector into a trained load monitoring model to obtain power load information of the key preset types of equipment; obtain equipment operation information of the key preset types of equipment and material feature change information in the production process of the cement production line, input the power load information, the equipment operation information and the material feature change information into a carbon emission monitoring model to obtain the total carbon emissions generated by the cement production line.

[0031] The above-mentioned carbon emission monitoring method, device, computer equipment, computer-readable storage medium and computer program product applied to the cement production line monitor the power load information of the cement production line equipment by using the load monitoring model, and monitor the carbon emissions generated by the cement production line by using the carbon emission monitoring model, and finally efficiently and accurately measure the total carbon emissions generated by the cement production line; thereby effectively solving the problems that the existing non-invasive monitoring method cannot meet the recognition accuracy, efficiency and robustness after being used in the cement production line, and the carbon emission monitoring method for cement production enterprises in the traditional technology is affected by high temperature, complex emission gas, low process continuity and large dynamic changes in the environment, resulting in low monitoring accuracy and difficult equipment maintenance. The problem, thereby improving the measurement accuracy of carbon emissions and realizing precise control and management of cement production line equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0033] FIG1 is an application environment diagram of a carbon emission monitoring method applied to a cement production line in one embodiment;

[0034] Figure 2 A schematic diagram of a process of a carbon emission monitoring method applied to a cement production line in one embodiment;

[0035] Figure 3 A schematic diagram of a flow chart of a data integration and sorting step in one embodiment;

[0036] Figure 4 A schematic diagram of a flow chart of a power load monitoring method in one embodiment;

[0037] Figure 5 A schematic diagram of a location for obtaining non-electrical characteristic data in one embodiment;

[0038] Figure 6 A schematic diagram of a network structure of an Adam-DBN load monitoring model in an embodiment;

[0039] Figure 7 A schematic diagram of a process for monitoring carbon emissions from cement production in an application embodiment;

[0040] Figure 8 A schematic diagram of a process for monitoring carbon emissions from cement production in another application embodiment;

[0041] Fig. 9 A schematic diagram of a process for training a carbon emission monitoring model in one embodiment;

[0042] Fig.10 is a structural block diagram of a carbon emission monitoring device applied to a cement production line in one embodiment;

[0043] FIG. 11 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] The carbon emission monitoring method for cement production line provided in the embodiment of the present application can be applied in the application environment as shown in FIG1. ​​The terminal communicates with the server through the network. The data storage system can store the data that the server needs to process. The data storage system can be integrated on the server, or placed on the cloud or other network servers.

[0046] Specifically, the carbon emission monitoring method applied to a cement production line provided in the embodiment of the present application can be executed by a server.

[0047] Exemplarily, the server obtains non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrates and sorts the non-electrical characteristic data, equipment status data and equipment power consumption data to obtain a target data set; the server constructs a joint feature vector corresponding to each non-electrical characteristic data according to the target data set; the server inputs the joint feature vector into a trained load monitoring model to obtain power load information of key preset types of equipment; the server obtains equipment operation information of key preset types of equipment and material characteristic change information in the production process of the cement production line, inputs the power load information, equipment operation information and material characteristic change information into the carbon emission monitoring model to obtain the total carbon emissions generated by the cement production line.

[0048] In the application environment shown in FIG1 , the terminal may be, but is not limited to, various personal computers, notebook computers, smart phones, and tablet computers. The server may be implemented as an independent server or a server cluster consisting of multiple servers.

[0049] In one embodiment, Figure 2 As shown, a carbon emission monitoring method applied to a cement production line is provided, and the method is applied to the server in FIG1 as an example for explanation, including the following steps:

[0050] Step S201, obtaining non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrating and sorting the non-electrical characteristic data, equipment status data and equipment power consumption data to obtain a target data set.

[0051] Among them, the key preset types of equipment in the cement production line can be divided into material processing equipment, thermal reaction and heat exchange equipment, material conveying equipment and environmental protection auxiliary equipment according to the working characteristics of each type of equipment; among them, material processing equipment may include crushers, grinders, packaging machines and roller presses, thermal reaction and heat exchange equipment may include preheaters, decomposition furnaces, rotary kilns and coolers, material conveying equipment may be conveyor belts, and environmental protection auxiliary equipment may include bag dust collectors and electrostatic precipitators.

[0052] Among them, non-electrical characteristic data include but are not limited to material speed V, gas temperature T, gas concentration PPM and gas flow Q, etc., equipment status data include but are not limited to open O, closed N and gear positions 01, 02 during operation, etc., equipment energy consumption includes but is not limited to current data A, voltage data V and power P, etc.

[0053] Specifically, in response to the carbon emission monitoring request of the cement production line, the server obtains the non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, cleans and integrates the non-electrical characteristic data, equipment status data and equipment power consumption data to obtain the target data set.

[0054] Step S202: constructing joint feature vectors corresponding to each non-electrical feature data according to the target data set.

[0055] The joint feature vector may be a feature vector including feature parameters, device state codes, and power consumption values.

[0056] Specifically, the server constructs the non-electrical characteristic data as a joint characteristic vector corresponding to the feed material flow rate, gas temperature and concentration, and gas flow rate, respectively, based on the target data set.

[0057] Step S203: input the joint feature vector into the trained load monitoring model to obtain the power load information of key preset types of equipment.

[0058] Among them, the load monitoring model can be an Adam-DBN load monitoring model, which is a power system load monitoring model that combines the Adam optimization algorithm and the Deep Belief Network (DBN).

[0059] Specifically, the server inputs the joint feature vector into the trained load monitoring model to perform parameter calculation to obtain the power load information of key preset types of equipment.

[0060] Step S204, obtaining equipment operation information of key preset types of equipment and material characteristic change information during the production process of the cement production line, inputting the power load information, equipment operation information and material characteristic change information into the carbon emission monitoring model, and obtaining the total carbon emissions generated by the cement production line.

[0061] The equipment operation information includes but is not limited to the temperature and rotation speed of the rotary kiln, the material characteristic change information may be the content ratio of each component of the raw material and the decomposition change, and the power load information may be the current, voltage and power.

[0062] Specifically, the server obtains the equipment operation information of key preset types of equipment and the material characteristic change information during the production process of the cement production line, inputs the power load information, equipment operation information and material characteristic change information into the carbon emission monitoring model for parameter calculation, and obtains the total carbon emissions generated by the cement production line.

[0063] The above-mentioned carbon emission monitoring method applied to the cement production line monitors the power load information of the cement production line equipment by using the load monitoring model, and monitors the carbon emissions generated by the cement production line by using the carbon emission monitoring model, and finally efficiently and accurately measures the total carbon emissions generated by the cement production line; thereby effectively solving the problems that the existing non-invasive monitoring methods cannot meet the recognition accuracy, efficiency and robustness after being used in the cement production line, and the carbon emission monitoring methods for cement production enterprises in traditional technologies are affected by high temperature, complex emission gases, low process continuity and large dynamic changes in the environment, resulting in low monitoring accuracy and difficult equipment maintenance. The problem, thereby improving the measurement accuracy of carbon emissions and realizing precise control and management of cement production line equipment.

[0064] In one embodiment, if Figure 3 As shown, in the above step S201, the non-electrical characteristic data, the device status data and the device power consumption data are integrated and sorted to obtain a target data set, which specifically includes the following steps:

[0065] Step S301 : integrating non-electrical characteristic data, device status data and device power consumption data at the same time interval.

[0066] Step S302 , arranging the integrated data in chronological order to form a current data set including a timestamp column and a corresponding indicator column.

[0067] Step S303, remove outliers and fill missing values ​​in the current data set to obtain a target data set.

[0068] Specifically, the server integrates non-electrical feature data, device status data, and device power consumption data at the same time interval; arranges the integrated data in chronological order to form a current data set containing a timestamp column and a corresponding indicator column; uses interpolation and KNN (K-Nearest Neighbors) algorithm to remove outliers and fill in missing values ​​in the current data set to obtain the target data set; in addition, the feature data, status data, and energy consumption data need to be standardized or normalized. After completing basic data processing, a joint feature vector needs to be constructed to facilitate model learning.

[0069] In this embodiment, the integrated data is arranged in chronological order to form a current data set including a timestamp column and a corresponding indicator column, and outliers are removed and missing values ​​are filled in the current data set, so as to quickly and accurately obtain the target data set and ensure the continuity and integrity of the data.

[0070] In one embodiment, if Figure 4 As shown, a load monitoring method based on adaptive motion optimization algorithm and deep belief network is provided, and the specific steps are as follows:

[0071] First of all, this solution combines the process flow of the entire production line and the characteristics of various equipment, aiming to systematically explore and integrate the non-electrical characteristic parameters of each key link, so as to obtain more comprehensive and accurate non-electrical characteristics that can reflect the operating status and energy consumption level of each equipment. According to the working characteristics of each type of equipment, all equipment on the production line is divided into material processing equipment, thermal reaction and heat exchange equipment, material conveying equipment and environmental protection auxiliary equipment. Among them, material processing equipment includes crushers, grinders, packaging machines and roller presses, thermal reaction and heat exchange equipment includes preheaters, decomposition furnaces, rotary kilns and coolers, material conveying equipment is a conveyor belt, and environmental protection auxiliary equipment includes bag dust collectors and electrostatic precipitators. By dividing all equipment, non-electrical characteristics can be selected more clearly and intuitively.

[0072] Secondly, for each type of equipment, select and formulate corresponding collection methods. The working status and power consumption of crushers, grinders and other equipment are directly related to the amount of material fed into them. The working status and power consumption of preheaters, decomposition furnaces, rotary kilns and coolers are directly related to the gas temperature and concentration in the dust collector. The working status and power consumption of the dust collector are directly related to the gas flow rate at the exhaust port. The amount of material fed can be reflected by measuring the flow rate of the material on the feeding conveyor belt through a radar speed meter. The material flow rate can also reflect the working status and power consumption of the conveyor belt. The gas temperature and concentration can be directly obtained through the monitoring data of the dust collector equipment; the gas flow rate at the exhaust port of the dust collector can be measured using an industrial standard flow meter.

[0073] Through consulting materials and investigation, the new dry process cement production technology has become the mainstream method of producing cement due to its high efficiency, high quality, energy saving, resource conservation, environmental protection and sustainable development. Therefore, this plan will combine the new dry process cement production line and analyze the specific implementation method of the above-mentioned method of collecting non-electrical characteristic parameters through a flow chart:

[0074] When collecting data, the specific location where the data is obtained is as follows Figure 5 As shown, v represents the measured material speed on the feeding conveyor belt; T and PPM represent the temperature and concentration of the internal gas monitored at the electric dust collector; Q represents the gas flow rate after purification at the dust collector exhaust port. The non-electrical characteristic data and the various power consumption data of the equipment measured by the smart meter and the synchronous records are set as the equipment status data for preprocessing. Each set of data includes three types of data: characteristic parameters, equipment status, and equipment energy consumption. The characteristic parameters include material speed V, gas temperature T, gas concentration PPM, and gas flow Q. The equipment status includes open O, closed N, and gears 01, 02 during operation. The equipment energy consumption includes current A, voltage V, and power P. The above three types of data are integrated at the same time interval, and the integrated data are arranged in chronological order to form a data set containing timestamp columns and corresponding indicator columns. In order to better train the model, interpolation and KNN algorithms are used to remove outliers and fill missing values ​​to ensure data continuity and integrity. In addition, the feature data, state data and energy consumption data need to be standardized or normalized. After completing the basic data processing, a joint feature vector needs to be constructed so that the model can learn. For each time step, a feature vector containing feature parameters, device state code and power consumption value can be constructed:

[0075] The characteristic vector that can be constructed for the equipment whose characteristic parameter is the feed material flow rate is:

[0076]

[0077] The characteristic vector that can be constructed for a device whose characteristic parameters are gas temperature and concentration is:

[0078]

[0079] The characteristic vector that can be constructed for a device whose characteristic parameter is gas flow is:

[0080]

[0081] The first element corresponds to the standardized characteristic parameter, the fifth to sixth elements are the switch state codes, the fourth to fourth-to-last elements are the gear codes (the length depends on how many gears there are actually. Here, it is assumed that there are only two gears, the switch state code is [1, 0], and the gear code is [0, 1, 0, ...] to represent the "02" gear), and the last three elements correspond to various standardized power consumption values.

[0082] Next, this solution involves a mathematical model based on an adaptive motion optimization algorithm and a deep belief network, referred to as the Adam-DBN load monitoring model. This model can effectively mine the deep nonlinear relationship and long- and short-term dependencies between non-electrical features and equipment power consumption, and can also accelerate the training process and optimize model performance through the Adam optimizer, so as to efficiently and accurately monitor production line equipment in real time.

[0083] Deep belief network (DBN) is a deep learning model composed of multiple restricted Boltzmann machines. DBN can be used for feature extraction and pattern recognition of data in monitoring applications, but as the RBM (multi-layer restricted Boltzmann machine) layer progresses, the speed of updating weights will become slow, and there is no clear direction for updating weights, which is easy to fall into the local optimal solution; the paper "Optical phased array output beam calibration method based on Adam algorithm" mentioned that the Adam algorithm achieves refinement and efficiency of parameter update by dynamically adjusting the learning rate of each parameter. This adaptive learning rate adjustment feature enables the Adam algorithm to effectively cope with the fluctuation of the evaluation function value, especially in complex optimization scenarios, which helps the algorithm avoid falling into the local optimal solution and get closer to the global optimal solution. Therefore, within the same number of iterations, the Adam algorithm is not only better than the SPDG algorithm and the GS algorithm in beam calibration quality, but also more efficient in optimization efficiency.

[0084] In one embodiment, using the Adam optimization algorithm to replace the BP algorithm in the DBN network can also help the DBN network find the minimum value of the loss function faster, improve training efficiency and convergence speed. Through the Adam algorithm, the network can more effectively update weights and biases, making the training process more stable and avoiding falling into a local optimal solution.

[0085] like Figure 6 As shown in the figure, the Adam-DBN load monitoring model is mainly divided into the pre-training stage and the tuning stage during model training. In the pre-training stage, firstly, the first RBM needs to be fully trained, the weight and offset of the first RBM are fixed, and then the state of its hidden neuron is used as the input vector of the second RBM. The above steps are repeated as many times as desired until the last layer. The number of RBM layers of the model here needs to be determined by repeated trials based on specific application scenarios and data characteristics; after the pre-training is completed, a supervised learning method is used for fine-tuning. At this time, the Adam optimizer can be used. In each iteration of the reverse tuning stage, the Adam optimization algorithm can calculate the loss function according to the forward propagation, and then calculate the gradient through back propagation, update the weights, and thus adjust the parameters of the entire network to minimize the prediction error.

[0086] Combined with the characteristics of cement production lines, this solution divides the preprocessed data set into a training set and a test set in a ratio of 9:1. Cement production lines face a variety of complex working conditions and potential failure modes, such as raw material quality fluctuations, equipment aging, and changes in environmental conditions. The 90% training set can cover more working conditions and abnormal situation samples, which helps the model learn a wider range of failure modes during the training process. In addition, the operation decisions of cement production lines have extremely high requirements for the accuracy and stability of model predictions. Any misjudgment may lead to reduced production efficiency, energy waste, or equipment damage. The 90% training set helps the model better learn the intrinsic structure of the data, reduce the risk of overfitting, and thus improve the prediction accuracy on unknown data. Retaining 10% as a test set can provide a large enough sample space for fair and effective evaluation of the performance of the model on unseen data, which is essential for frequent model verification, tuning, and updating in the cement production line environment.

[0087] In one embodiment, if Figure 7 As shown, a method for establishing a prediction model for carbon emissions from cement production based on material characteristics and equipment power load information is provided. Through the changes in material characteristics, equipment power load information and the operating status of some equipment during the production process, a learning model is established to realize carbon emission monitoring in the cement production process. The specific contents are as follows:

[0088] During the raw material powder and raw material preparation stage, electricity consumption emissions are mainly generated, including crushers, raw material mills, and homogenization storage tanks. The raw materials after homogenization enter the firing stage. The emissions in this stage include electricity consumption emissions and direct decomposition emissions. After this stage, the raw materials become clinker after a series of changes. After the clinker processing is completed, it enters the cooling and fine grinding stage. This stage mainly generates electricity consumption emissions. After this stage, the cement processing is completed. The processed cement mainly generates energy consumption emissions during transportation. Since the production and transportation links of cement enterprises are usually managed and monitored separately, the carbon emission monitoring of the cement transportation process is usually carried out independently from the carbon emission monitoring of the cement production process. This plan mainly involves carbon emission monitoring in the cement production link, and does not involve carbon emission monitoring in the transportation link after cement processing is completed. In the cement production link, such as Figure 8 As shown, the total carbon emissions generated is EC= (n=5). As mentioned above, the stage most affected by materials and the environment is the clinker burning stage. Therefore, during the transportation of raw materials, spectral imaging technology can be used to analyze the composition of the entire material flow to obtain the content ratio of each component of the raw material, that is, the material characteristics. After obtaining the characteristic information of the raw material, it is also necessary to obtain the processing information, that is, the decomposition and change of the raw material. As one of the core equipment for cement production, the temperature and speed of the rotary kiln can well reflect the decomposition and change of the raw material. Since the operating state of the rotary kiln is controllable, its temperature and speed can be obtained from the operating terminal, and the two parameters can be transmitted to the prediction model at regular intervals. After completing the collection of material characteristics and the temperature and speed of the rotary kiln, combined with the power consumption information of the equipment, the carbon emissions in the cement production process can be calculated through the established model.

[0089] In one embodiment, if Fig. 9 As shown in the figure, the power consumption of each device and the material characteristics of spectral imaging technology obtained by the Adam-DBN load monitoring model are used as inputs. At the same time, the temperature and speed of the rotary kiln are used as inputs again. The material feature dimension N is defined and the merging layer is set. In order to learn the relationship between carbon emissions and material and power consumption, two fully connected layers are selected, and the output layer is set to carbon dioxide emissions. After the model is trained, the electric carbon material model is established. Three evaluation indicators are used to measure the prediction performance of the model, namely the mean square error (MSE), the mean absolute error (MAE) and the determination coefficient (R 2 ). Smaller MSE and MAE values ​​indicate higher estimation accuracy because they reflect the ability of the model to closely match the true value. 2 It usually ranges from 0 to 1, where values ​​closer to 1 indicate a better fit, indicating that the model explains a greater proportion of the variability in the data. After the model reaches the target performance, it can predict the carbon emissions of cement production enterprises through material characteristics, total electricity consumption, rotary kiln speed and temperature.

[0090] In one embodiment, a carbon emission monitoring method applied to a cement production line in a specific embodiment is provided, which specifically includes the following steps:

[0091] S1. Arrange the raw material spectrum sampling system to realize timed sampling in different time periods and record the sampling time.

[0092] S2. Use filtering technology or wavelet transform to remove noise from spectral data.

[0093] S3. Input the processed data into a pre-trained model. In this example, a regression model based on a support vector machine is selected to predict the content of each component in the cement raw material and save the characteristic information of the sampled raw material.

[0094] S4. Add smart meters to the power supply system of cement enterprises to monitor the total power consumption and save the total power consumption according to the raw material sampling time.

[0095] S5. Obtain the temperature and speed information of the rotary kiln from the operation terminal of the production line.

[0096] S6. Preprocess the obtained material characteristic data, total power consumption data, rotary kiln speed and temperature data to make these data into a form that can be input into the model. In this example, the time series length is 10, the power consumption characteristic dimension is 1, the rotary kiln state dimension is 2, and the number of equipment is 7.

[0097] S7. One-stage model training. The training data set is the historical electricity consumption data of the enterprise and the historical working status of the rotary kiln. The power consumption of each device is obtained using the Adam-DBN load monitoring model.

[0098] S8. Take the output of the first stage - the power consumption of each equipment, the temperature and speed of the rotary kiln, and the proportion of raw material as input.

[0099] S9. Second-stage model training. The training data set is the historical carbon emission records of the enterprise.

[0100] S10. Three evaluation indicators are used to measure the prediction performance of the model, namely, mean square error (MSE), mean absolute error (MAE) and coefficient of determination (R 2 ). Smaller MSE and MAE values ​​indicate higher estimation accuracy because they reflect the ability of the model to closely match the true value. 2 Typically ranges from 0 to 1, where values ​​closer to 1 indicate a better fit, indicating that the model explains a greater proportion of the variability in the data.

[0101] After the model reaches the target performance, it can predict the carbon emissions of cement production enterprises through material characteristics, total electricity consumption, rotary kiln speed and temperature.

[0102] Compared with the existing technical solutions, this technical solution has the following beneficial effects:

[0103] 1. Non-electrical features can be selected according to the characteristics of the equipment. They are directly related to the operating status of the equipment and can better reflect the actual status of the equipment, thereby improving the prediction accuracy. Compared with electrical features, non-electrical features are relatively stable and not easily affected by the external environment, thereby reducing interference factors and making the model output results more accurate.

[0104] 2. Compared with the previous DBN model, the new model obtained by optimizing the DBN model using the Adam algorithm has stronger training efficiency, robustness and generalization ability when facing the complexity, noise and potential sparse gradient problems of cement production line data, and can build a more accurate and stable equipment status prediction and power consumption estimation model. In the case that there may be a large number of sparse features or sparse gradient changes in cement production line data, it can better cope with these situations, avoid learning falling into local minima or over-reliance on certain sparse features, and has better robustness; in addition, because Adam can quickly capture the global trend of the loss function and effectively adjust the learning rate, the model can achieve lower training loss in fewer iterations, improving the training efficiency of the model while maintaining good generalization performance.

[0105] 3. Compared with the traditional electric carbon model, this solution adds equipment operating status and material characteristics into the calculation range to improve the reliability of model prediction.

[0106] 4. Compared with the traditional method of using model prediction, this solution is universal for cement production enterprises in different environments, because the input required by the model of this solution is the root cause of carbon emissions and is not affected by environmental factors.

[0107] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0108] Based on the same inventive concept, the embodiment of the present application also provides a carbon emission monitoring device applied to a cement production line for implementing the carbon emission monitoring method applied to a cement production line involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more embodiments of the carbon emission monitoring device applied to a cement production line provided below can refer to the limitations of the carbon emission monitoring method applied to a cement production line above, and will not be repeated here.

[0109] In an exemplary embodiment, Fig.10 As shown, a carbon emission monitoring device applied to a cement production line is provided, comprising:

[0110] The data processing module 1001 is used to obtain non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical characteristic data, equipment status data and equipment power consumption data to obtain a target data set;

[0111] A vector construction module 1002 is used to construct a joint feature vector corresponding to each non-electrical feature data according to the target data set;

[0112] The information acquisition module 1003 is used to input the joint feature vector into the trained load monitoring model to obtain the power load information of key preset types of equipment;

[0113] The carbon emission monitoring module 1004 is used to obtain the equipment operation information of key preset types of equipment and the material characteristic change information during the production process of the cement production line, input the power load information, equipment operation information and material characteristic change information into the carbon emission monitoring model, and obtain the total carbon emissions generated by the cement production line.

[0114] In one embodiment, the data processing module 1001 is also used to integrate non-electrical characteristic data, device status data and device power consumption data at the same time interval; arrange the integrated data in chronological order to form a current data set including a timestamp column and a corresponding indicator column; and remove outliers and fill missing values ​​in the current data set to obtain a target data set.

[0115] In one embodiment, the carbon emission monitoring device applied to a cement production line also includes a first training module, which is used to divide the target data set into a training set and a test set according to a preset ratio; input the training set into the load monitoring model to be trained for iterative training to obtain a preliminarily trained load monitoring model; use the test set to test and tune the preliminarily trained load monitoring model until the preliminarily trained load monitoring model meets the accuracy and stability conditions, thereby obtaining a trained load monitoring model.

[0116] In one embodiment, the first training module is also used to train the first neural network model using the training set and determine the weight and offset of the first neural network model; obtain the state information of the hidden neurons of the first neural network model based on the weight and offset, use the state information as the input vector of the second neural network model, and tune the second neural network model based on the input vector; use the preset optimizer to adjust the first neural network model and the second neural network model to obtain a preliminarily trained load monitoring model.

[0117] In one embodiment, the carbon emission monitoring device applied to a cement production line also includes a second training module, which is used to construct an initial carbon emission monitoring model including an input layer, a merging layer, a fully connected layer and an output layer; based on the power load information, the equipment operation information and the material characteristic change information, the target mapping relationship between the carbon emissions and the power load information, the equipment operation information and the material characteristic change information is determined; according to the target mapping relationship, the model parameters of the initial carbon emission monitoring model are adjusted to obtain a carbon emission monitoring model.

[0118] In one embodiment, the carbon emission monitoring device applied to a cement production line also includes a model adjustment module, which is used to obtain the evaluation value of the carbon emission monitoring model under multiple preset evaluation indicators, and determine whether the evaluation value meets the corresponding threshold conditions; if any evaluation value does not meet the threshold conditions, determine the target model parameters corresponding to the current evaluation indicator, and adjust the target model parameters until the evaluation value of the current evaluation indicator meets the threshold conditions.

[0119] Each module in the carbon emission monitoring device for cement production line can be implemented in whole or in part by software, hardware and their combination. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0120] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG11. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a carbon emission monitoring method applied to a cement production line is implemented.

[0121] Those skilled in the art will understand that the structure shown in FIG. 11 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0122] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0123] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0124] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0125] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0126] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0127] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0128] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A carbon emission monitoring method for cement production line, It is characterized in that The method comprises: Acquire non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical characteristic data, the equipment status data and the equipment power consumption data to obtain a target data set; According to the target data set, constructing joint feature vectors corresponding to each of the non-electrical feature data; Inputting the joint feature vector into the trained load monitoring model to obtain the power load information of the key preset type of equipment; The equipment operation information of the key preset types of equipment and the material characteristic change information during the production process of the cement production line are obtained, and the power load information, the equipment operation information and the material characteristic change information are input into a carbon emission monitoring model to obtain the total carbon emissions generated by the cement production line.

2. The method according to claim 1, It is characterized in that The integrating and sorting the non-electrical characteristic data, the device status data and the device power consumption data to obtain a target data set includes: Integrating the non-electrical characteristic data, the device status data and the device power consumption data at the same time interval; Arrange the integrated data in chronological order to form a current data set containing a timestamp column and a corresponding indicator column; Outliers are removed and missing values ​​are filled in the current data set to obtain the target data set.

3. The method according to claim 1, It is characterized in that Before inputting the joint feature vector into the trained load monitoring model, the method further includes: Dividing the target data set into a training set and a test set according to a preset ratio; Inputting the training set into the load monitoring model to be trained for iterative training to obtain a preliminarily trained load monitoring model; The test set is used to test and tune the initially trained load monitoring model until the initially trained load monitoring model meets accuracy and stability conditions, thereby obtaining the trained load monitoring model.

4. The method according to claim 3, It is characterized in that The step of inputting the training set into the load monitoring model to be trained for iterative training to obtain a preliminarily trained load monitoring model includes: Using the training set, training a first neural network model, and determining weights and offsets of the first neural network model; According to the weight and the offset, the state information of the hidden neurons of the first neural network model is obtained, the state information is used as the input vector of the second neural network model, and the second neural network model is tuned according to the input vector; The first neural network model and the second neural network model are adjusted using a preset optimizer to obtain the preliminarily trained load monitoring model.

5. The method according to claim 1, It is characterized in that Before inputting the power load information, the equipment operation information and the material characteristic change information into the carbon emission monitoring model, the method further includes: Construct an initial carbon emission monitoring model including input layer, merging layer, fully connected layer and output layer; Based on the power load information, the equipment operation information and the material characteristic change information, determining a target mapping relationship between carbon emissions and the power load information, the equipment operation information and the material characteristic change information; According to the target mapping relationship, the model parameters of the initial carbon emission monitoring model are adjusted to obtain the carbon emission monitoring model.

6. The method according to claim 5, It is characterized in that The method further comprises: Obtaining evaluation values ​​of the carbon emission monitoring model under multiple preset evaluation indicators, and determining whether the evaluation values ​​meet corresponding threshold conditions; In the case that any of the evaluation values ​​does not meet the threshold condition, a target model parameter corresponding to the current evaluation indicator is determined, and the target model parameter is adjusted until the evaluation value of the current evaluation indicator meets the threshold condition.

7. A carbon emission monitoring device for cement production line, It is characterized in that The device comprises: A data processing module, used to obtain non-electrical characteristic data, equipment status data and equipment power consumption data of key preset types of equipment in the cement production line, integrate and sort the non-electrical characteristic data, the equipment status data and the equipment power consumption data to obtain a target data set; A vector construction module, used for constructing joint feature vectors corresponding to each of the non-electrical feature data according to the target data set; An information acquisition module, used for inputting the joint feature vector into the trained load monitoring model to obtain the power load information of the key preset type of equipment; The carbon emission monitoring module is used to obtain the equipment operation information of the key preset types of equipment and the material characteristic change information during the production process of the cement production line, input the power load information, the equipment operation information and the material characteristic change information into the carbon emission monitoring model, and obtain the total carbon emissions generated by the cement production line.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.