Electrolytic aluminum electric carbon portrait construction method and device based on improved self-organizing network

Abstract

The invention discloses an electrolytic aluminum electric carbon portrait construction method and device based on an improved self-organizing network, and the method comprises the steps: obtaining power load data in real time from each production link through setting a data collection strategy; constructing an electrolytic aluminum load model according to the power load data; dividing the electrolytic aluminum load into set production states through the electrolytic aluminum load model; according to the power load data, an enterprise carbon quota model and an actual measurement carbon emission model are combined to construct an electrolytic aluminum production process carbon emission model; quantifying the carbon emission flow condition of each production link, and obtaining the electrical carbon characteristics in a set production state; inputting the electric carbon features into an SOM network based on a sparrow optimization algorithm for iterative clustering processing, outputting an optimal clustering result, and generating an electric carbon portrait; and the production links in the high-carbon emission state are analyzed and obtained, and load scheduling and production strategy optimization are carried out on the production links. According to the method, refined carbon emission strategies for different production states can be generated, and the utilization efficiency of the carbon quota is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of high-energy-consuming industrial electric-carbon coupling technology, and in particular to a method and device for constructing an electric-carbon image of electrolytic aluminum based on an improved self-organizing network. Background Art

[0002] The electrolytic aluminum production process is characterized by high energy consumption and high carbon emissions, and the production load fluctuates with market demand, production plans, electricity prices and other factors. At the same time, electrolytic aluminum enterprises are also facing the dual pressures of carbon quota restrictions and price fluctuations in the electricity market, which further increases the complexity and uncertainty of enterprises in energy use and carbon emission management. Traditional management models and technical means are difficult to adapt to these changing demands, resulting in enterprises facing high carbon costs and unnecessary electricity consumption. In order to achieve green production, reduce costs and increase efficiency, electrolytic aluminum enterprises are in urgent need of innovative technical solutions to optimize load scheduling and carbon emission management, so as to achieve the goals of improving energy efficiency, reducing carbon emissions and achieving sustainable development.

[0003] At present, in view of the complex and changeable load characteristics and carbon emission conditions of electrolytic aluminum enterprises, few studies have been conducted to improve energy efficiency and reduce carbon emissions by constructing electric carbon portraits. Existing electric carbon portrait construction schemes mainly rely on the Self-Organizing Map (SOM) algorithm and other traditional clustering analysis methods to identify power load patterns and preliminarily outline user portraits. However, the traditional SOM algorithm is highly dependent on manual tuning and cannot achieve dynamic adjustment. Under the variable load and carbon emission environment of electrolytic aluminum enterprises, static parameter selection is difficult to adapt to real-time load scheduling needs, which directly affects the optimal utilization of carbon quotas. In addition, most existing schemes rely only on load curves for clustering, which makes it difficult to comprehensively consider the multi-dimensional carbon emission characteristics of electrolytic aluminum enterprises. Such a single feature analysis leads to insufficient carbon management accuracy and cannot provide refined carbon emission strategies for different production states, thereby reducing the utilization efficiency of carbon quotas and limiting the promotion of low-carbon production.

[0004] Therefore, how to invent a method for constructing an electric carbon portrait of an electrolytic aluminum enterprise that can generate refined carbon emission strategies for different production states and improve the utilization efficiency of carbon quotas has become an urgent problem to be solved. Summary of the invention

[0005] To this end, the present invention provides a method and device for constructing an electrolytic aluminum carbon portrait based on an improved self-organizing network, which can generate refined carbon emission strategies for different production states, improve the utilization efficiency of carbon quotas, enable enterprises to clearly understand the carbon emission characteristics under various load states, and provide data support for carbon emission responsibility management and optimal use of carbon quotas.

[0006] In order to achieve the above object, the present invention provides the following technical solution: a method for constructing electrolytic aluminum carbon image based on an improved self-organizing network, comprising:

[0007] By setting a data collection strategy, real-time power load data is obtained from each production link; the power load data includes the power consumption and production load status of the electrolytic aluminum equipment;

[0008] According to the power load data, an electrolytic aluminum load model is constructed; through the electrolytic aluminum load model, the electrolytic aluminum load is divided into set production states; the set production states include: rated production state, reduced production state, insulation state and cooling state;

[0009] According to the power load data, combined with the enterprise carbon quota model and the actual measured carbon emission model, a carbon emission model for the electrolytic aluminum production process is constructed; the carbon emission flow of each production link is quantified through the carbon emission model for the electrolytic aluminum production process, and the electric carbon characteristics under the set production state are obtained;

[0010] Inputting the electric carbon characteristics into a SOM network based on a sparrow optimization algorithm for iterative clustering processing, and outputting an optimal clustering result; generating an electric carbon portrait reflecting the characteristics of the set production state according to the optimal clustering result;

[0011] According to the electric-carbon portrait, the production links with high carbon emission status are analyzed and obtained; and load scheduling and production strategy optimization are performed on the production links with high carbon emission status.

[0012] As a preferred solution of the electrolytic aluminum electric carbon image construction method based on the improved self-organizing network, the expression of the electrolytic aluminum load model is:

[0013]

[0014] P l,min ≤P l,t ≤P l,max

[0015] Where P l,t I is the power of the electrolytic aluminum load l at time t; l,t is the current intensity of the electrolytic aluminum load l at time t; R l,t is the equivalent resistance; E l,t is the equivalent electromotive force; I max and I min The upper and lower limits of the current allowed to pass into the electrolytic aluminum load l; P l,max and P l,min They are the maximum and minimum values ​​of the electrolytic aluminum load l respectively.

[0016] As a preferred solution for the method of constructing electrolytic aluminum carbon profile based on improved self-organizing network, the expression of the enterprise carbon quota model is:

[0017]

[0018] In the formula, E p Total carbon allowances obtained for all electrolytic aluminium loads; E p,lt is the carbon emission quota obtained by the electrolytic aluminum load l at time t; k is the electricity consumption per ton of aluminum; P is the carbon emission quota value obtained per unit of electricity electrolytic aluminum load l; lt is the power of the electrolytic aluminum load l at time t.

[0019] As a preferred solution of the electrolytic aluminum carbon image construction method based on the improved self-organizing network, the expression of the actual measured carbon emission model is:

[0020] E=E b +E m +E PFC +E e

[0021] Where, E is the total greenhouse gas emissions of the enterprise; E m Carbon emissions caused by the use of energy as raw materials; E PFC is the carbon emissions in the industrial production process; E e Carbon emissions resulting from net purchased electricity consumption.

[0022] As a preferred solution of the method for constructing electrolytic aluminum electrocarbon portrait based on improved self-organizing network, in the process of inputting the electrocarbon features into the SOM network based on sparrow optimization algorithm for iterative clustering processing, the iterative clustering processing steps are:

[0023] Initialize the population parameters of the sparrow optimization algorithm;

[0024] Input the user characteristic data into the SOM network represented by individual sparrows;

[0025] Dynamically adjust the learning rate, neighborhood function and weight of the SOM network;

[0026] Input the user feature data into the SOM network after the parameters are dynamically adjusted, perform iterative adjustment until the set number of iterations is reached, and output the SOM clustering results and clustering indicators;

[0027] Optimizing the population parameters of the sparrow algorithm according to the clustering index to obtain an optimized population;

[0028] Based on the optimized population, an optimal clustering result is output; based on the optimal clustering result, an electro-carbon portrait reflecting the characteristics of the set production state is generated.

[0029] The present invention also provides an electrolytic aluminum electrocarbon image construction device based on an improved self-organizing network, and the electrolytic aluminum electrocarbon image construction method based on the improved self-organizing network includes:

[0030] The power load data acquisition module is used to acquire power load data from each production link in real time by setting a data acquisition strategy; the power load data includes the power consumption and production load status of the electrolytic aluminum equipment;

[0031] An electrolytic aluminum load model building module is used to build an electrolytic aluminum load model according to the power load data; through the electrolytic aluminum load model, the electrolytic aluminum load is divided into set production states; the set production states include: rated production state, reduced production state, insulation state and cooling state;

[0032] The module for constructing the carbon emission model of the electrolytic aluminum production process is used to construct the carbon emission model of the electrolytic aluminum production process according to the power load data, in combination with the enterprise carbon quota model and the actual measured carbon emission model; the carbon emission flow of each production link is quantified through the carbon emission model of the electrolytic aluminum production process, and the electric carbon characteristics under the set production state are obtained;

[0033] An electric carbon image generation module is used to input the electric carbon characteristics into a SOM network based on a sparrow optimization algorithm for iterative clustering processing, and output an optimal clustering result; and generate an electric carbon image reflecting the characteristics of the set production state according to the optimal clustering result;

[0034] The application and optimization module is used to analyze and obtain the production links with high carbon emission status according to the electric carbon portrait; and to perform load scheduling and production strategy optimization on the production links with high carbon emission status.

[0035] As a preferred solution of the electrolytic aluminum electric carbon image construction device based on the improved self-organizing network, in the electrolytic aluminum load model construction module, the expression of the electrolytic aluminum load model is:

[0036]

[0037] P l,min ≤P l,t ≤P l,max

[0038] Where P l,t I is the power of the electrolytic aluminum load l at time t; l,t is the current intensity of the electrolytic aluminum load l at time t; R l,t is the equivalent resistance; El,t is the equivalent electromotive force; I max and I min The upper and lower limits of the current allowed to pass into the electrolytic aluminum load l; P l,max and P l,min They are the maximum and minimum values ​​of the electrolytic aluminum load l respectively.

[0039] As a preferred solution of the electrolytic aluminum electric carbon image construction device based on the improved self-organizing network, in the electrolytic aluminum production process carbon emission model construction module, the expression of the enterprise carbon quota model is:

[0040]

[0041] In the formula, E p Total carbon allowances obtained for all electrolytic aluminium loads; E p,lt is the carbon emission quota obtained by the electrolytic aluminum load l at time t; k is the electricity consumption per ton of aluminum; P is the carbon emission quota value obtained per unit of electricity electrolytic aluminum load l; lt is the power of the electrolytic aluminum load l at time t.

[0042] As a preferred solution of the electrolytic aluminum carbon image construction device based on the improved self-organizing network, in the electrolytic aluminum production process carbon emission model construction module, the expression of the actual measured carbon emission model is:

[0043] E=E b +E m +E PFC +E e

[0044] Where, E is the total greenhouse gas emissions of the enterprise; E m Carbon emissions caused by the use of energy as raw materials; E PFC is the carbon emissions in the industrial production process; E e Carbon emissions resulting from net purchased electricity consumption.

[0045] As a preferred solution of the electrolytic aluminum carbon image construction device based on the improved self-organizing network, in the carbon image generation module, the iterative clustering processing submodule includes:

[0046] The parameter initialization submodule is used to initialize the population parameters of the sparrow optimization algorithm;

[0047] A feature data input submodule is used to input user feature data into the SOM network represented by the individual sparrows;

[0048] The SOM network parameter dynamic adjustment submodule is used to dynamically adjust the learning rate, neighborhood function and weight of the SOM network;

[0049] The SOM network iterative processing submodule is used to input the user feature data into the SOM network after the parameters are dynamically adjusted, perform iterative adjustment until the set number of iterations is reached, and output the SOM clustering results and clustering indicators;

[0050] A population parameter optimization submodule, used to optimize the population parameters of the sparrow algorithm according to the clustering index to obtain an optimized population;

[0051] The electric carbon portrait generation submodule is used to output the optimal clustering result according to the optimized population; and generate the electric carbon portrait reflecting the characteristics of the set production state according to the optimal clustering result.

[0052] The present invention has the following advantages: by setting a data collection strategy, power load data is obtained from each production link in real time; the power load data includes the power consumption and production load status of the electrolytic aluminum equipment; according to the power load data, an electrolytic aluminum load model is constructed; through the electrolytic aluminum load model, the electrolytic aluminum load is divided into set production states; the set production states include: rated production state, reduced production state, insulation state and cooling state; according to the power load data, combined with the enterprise carbon quota model and the actual measurement carbon emission model, a carbon emission model of the electrolytic aluminum production process is constructed; through the carbon emission model of the electrolytic aluminum production process, the carbon emission flow of each production link is quantified to obtain the electric carbon characteristics under the set production state; the electric carbon characteristics are input into the SOM network based on the sparrow optimization algorithm for iterative clustering processing, and the optimal clustering result is output; according to the optimal clustering result, an electric carbon portrait reflecting the characteristics of the set production state is generated; according to the electric carbon portrait, the production link with high carbon emission state is analyzed and obtained; load scheduling and production strategy optimization are performed on the production link with high carbon emission state. The present invention combines the load model, free carbon quota and actual carbon emission model of the electrolytic aluminum enterprise, and constructs a multi-dimensional electricity-carbon portrait through carbon emission flow analysis. The portrait integrates the characteristics of electricity use and carbon emission, and divides production states with similar carbon emissions into the same group, helping enterprises to accurately manage carbon emission responsibilities under various load conditions, optimize the use of carbon quotas, and thus effectively improve energy efficiency and promote low-carbon production. In view of the complex and changeable load and carbon emission characteristics of electrolytic aluminum enterprises, the present invention proposes an improved self-organizing map neural network (ISOM) algorithm combined with the sparrow optimization algorithm (SSA). The algorithm overcomes the dependence of the traditional SOM algorithm on manual parameter tuning, and through dynamic parameter adjustment, it better adapts to the real-time load scheduling needs of electrolytic aluminum enterprises under different production conditions, thereby improving the utilization efficiency of carbon quotas and realizing accurate load and carbon emission optimization management. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0054] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0055] Figure 1 A schematic flow chart of a method for constructing an electrolytic aluminum carbon profile based on an improved self-organizing network provided in Example 1 of the present invention;

[0056] Figure 2 A schematic diagram of a specific implementation process of the method for constructing an electrolytic aluminum carbon profile based on an improved self-organizing network provided in Example 1 of the present invention;

[0057] Figure 3 This is a schematic diagram of the architecture of the device for constructing the electrolytic aluminum carbon image based on the improved self-organizing network provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0058] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Example 1

[0060] See also Figure 1 and Figure 2 Embodiment 1 of the present invention provides a method for constructing an electrolytic aluminum carbon image based on an improved self-organizing network, comprising the following steps:

[0061] S1. By setting a data collection strategy, real-time power load data is obtained from each production link; the power load data includes power consumption and production load status of electrolytic aluminum equipment;

[0062] S2. constructing an electrolytic aluminum load model according to the power load data; dividing the electrolytic aluminum load into set production states through the electrolytic aluminum load model; the set production states include: rated production state, reduced production state, heat preservation state and cooling state;

[0063] S3. Based on the power load data, combined with the enterprise carbon quota model and the actual measured carbon emission model, a carbon emission model for the electrolytic aluminum production process is constructed; the carbon emission flow of each production link is quantified through the carbon emission model for the electrolytic aluminum production process to obtain the electric carbon characteristics under the set production state;

[0064] S4, inputting the electric carbon characteristics into the SOM network based on the sparrow optimization algorithm for iterative clustering processing, and outputting the optimal clustering result; generating an electric carbon portrait reflecting the characteristics of the set production state according to the optimal clustering result;

[0065] S5. Analyze and obtain the production links with high carbon emission status according to the electric-carbon portrait; and perform load scheduling and production strategy optimization on the production links with high carbon emission status.

[0066] In this embodiment, in step S1, by setting a data collection strategy, power load data is acquired from each production link in real time; the power load data includes power consumption and production load status of electrolytic aluminum equipment;

[0067] Specifically, through the enterprise power data collection system, real-time power load data is obtained from various production links within the enterprise, including the power consumption and production load status of electrolytic aluminum equipment.

[0068] In this embodiment, in step S2, an electrolytic aluminum load model is constructed based on the power load data; the electrolytic aluminum load is divided into set production states through the electrolytic aluminum load model; the set production states include: rated production state, reduced production state, insulation state and cooling state.

[0069] Specifically, according to the power load data, an electrolytic aluminum load model is constructed; according to the current value, the electrolytic aluminum load can be divided into set production states: rated production state, reduced production state, insulation state and cooling state.

[0070] Wherein, the expression of the electrolytic aluminum load model is:

[0071]

[0072] P l,min ≤P l,t ≤P l,max

[0073] Where P l,tI is the power of the electrolytic aluminum load l at time t; l,t is the current intensity of the electrolytic aluminum load l at time t; R l,t is the equivalent resistance; E l,t is the equivalent electromotive force; I max and I min The upper and lower limits of the current allowed to pass into the electrolytic aluminum load l; P l,max and P l,min They are the maximum and minimum values ​​of the electrolytic aluminum load l respectively.

[0074] In this embodiment, in step S3, a carbon emission model of the electrolytic aluminum production process is constructed according to the power load data, combined with the enterprise carbon quota model and the actual measured carbon emission model; the carbon emission flow of each production link is quantified through the carbon emission model of the electrolytic aluminum production process, and the electric carbon characteristics under the set production state are obtained;

[0075] Specifically, by analyzing the collected load data, evaluating the electricity usage of each production link, and combining the company's carbon quota model and actual measurement carbon emission model, a carbon emission model for the electrolytic aluminum production process is constructed. The carbon emission model for the electrolytic aluminum production process quantifies the carbon emission flow of each link, thereby revealing the electricity-carbon characteristics of the company under different load conditions.

[0076] The expression of the enterprise carbon quota model is:

[0077]

[0078] In the formula, E p Total carbon allowances obtained for all electrolytic aluminium loads; E p,lt is the carbon emission quota obtained by the electrolytic aluminum load l at time t; k is the electricity consumption per ton of aluminum; P is the carbon emission quota value obtained per unit of electricity electrolytic aluminum load l; lt is the power of the electrolytic aluminum load l at time t.

[0079] In this embodiment, the total greenhouse gas emissions of the electrolytic aluminum load are equal to the sum of the emissions from the use of energy as raw materials in the aluminum electrolysis process, the emissions from the industrial production process, and the emissions from the net purchased electricity consumption of the enterprise.

[0080] Specifically, the expression of the actual measured carbon emission model is:

[0081] E=E b +E m +E PFC +E e

[0082] Where, E is the total greenhouse gas emissions of the enterprise; E mCarbon emissions caused by the use of energy as raw materials; E PFC is the carbon emissions in the industrial production process; E e Carbon emissions resulting from net purchased electricity consumption.

[0083] Among them, the carbon emissions caused by the use of energy as a raw material are mainly attributed to the continuous consumption of carbon anodes as reducing agents in chemical reactions, thereby releasing carbon dioxide. The calculation formula for carbon emissions caused by the use of energy as a raw material is:

[0084]

[0085] In the formula, E m E is the total carbon emissions caused by the consumption of raw materials by electrolytic aluminum enterprises; m,lt is the carbon emission caused by raw material consumption of the lth electrolytic cell series at time t; EF c is the carbon dioxide emission factor consumed in the carbon anode.

[0086] The calculation formula for the carbon dioxide emission factor consumed by the carbon anode is:

[0087]

[0088] In the formula, NC c A is the net consumption of aluminum carbon anode per ton; c is the average ash content of carbon anode; S c is the average sulfur content of the carbon anode.

[0089] In this embodiment, during the production of electrolytic aluminum, factors such as low electrolyte levels or delayed processing time may cause the anode effect. At this time, not only will the cell voltage increase sharply, but perfluorocarbons CF4 and C2F6 will also be emitted. The calculation formula for greenhouse gas emissions caused by the anode effect is:

[0090]

[0091] In the formula, E PFC is the total amount of carbon emissions caused by industrial production processes; is the CF4 emission factor of the anode effect; is the C2F6 emission factor of the anode effect.

[0092] In this embodiment, in order to ensure the normal production of electrolytic aluminum and the stable operation of equipment within the enterprise, the electrolytic aluminum enterprise usually purchases additional electricity.

[0093] The formula for calculating carbon emissions caused by net purchased electricity consumption is:

[0094] E e =EF e ·Wne

[0095] In the formula, E e Carbon emissions caused by net purchase of electricity for aluminum smelters; EF e is the emission factor of electricity consumption; W ne is the electricity consumption.

[0096] In this embodiment, in step S4, the electric carbon characteristics are input into the SOM network based on the sparrow optimization algorithm for iterative clustering processing, and the optimal clustering result is output; and an electric carbon portrait reflecting the characteristics of the set production state is generated according to the optimal clustering result;

[0097] Specifically, the present invention uses an improved self-organizing map network (ISOM) for cluster analysis, and inputs the extracted electric carbon features into a SOM network based on a sparrow optimization algorithm (SSA) for automatic clustering. The improved SOM algorithm can classify production states with similar power usage characteristics and carbon emissions into the same group through a triangular topology neighborhood structure and automatic parameter tuning. Through clustering, the load mode and carbon emission mode under different production states can be identified, thereby forming a representative electric carbon portrait.

[0098] The iterative clustering processing steps are:

[0099] S41, initializing the population parameters of the sparrow optimization algorithm;

[0100] Specifically, the population parameters of the sparrow optimization algorithm are initialized, and each sparrow individual contains different SOM node initial weights, learning rates, and network topology structures.

[0101] S42, inputting the user feature data into the SOM network represented by the individual sparrows;

[0102] Specifically, the user feature data is input into the SOM of each individual representative, and the similarity d j Find the best node j in the competitive layer:

[0103] j = argmin{d j (x)}=argmin||xw j ||

[0104] In the formula, ||·|| represents the Euclidean distance; x represents the input feature vector; w j Represents the weight vector of neuron j in the competition layer.

[0105] S43, dynamically adjusting the learning rate, neighborhood function and weight of the SOM network;

[0106] Specifically,

[0107] Δwi (k) = η1(k)·G j (k,i)·(x(k)-w i (k))

[0108]

[0109]

[0110] In the formula, Δw i (k) represents the weight update value of the i-th competitive layer neuron in the k-th iteration; η1(k) represents the weight learning rate in the k-th iteration; η w is the initial weight learning rate, G j (k,i) represents the triangular topology neighborhood function of the weight of node i near the winning node j in the competition layer, σ w Represents the weight neighborhood radius.

[0111] S44, inputting the user feature data into the SOM network after the parameters are dynamically adjusted, performing iterative adjustment until the set number of iterations is reached, and outputting the SOM clustering results and clustering indicators;

[0112] Specifically, step S42 and step S43 are repeated until the set maximum number of iterations is reached, and the SOM clustering result and clustering index are output.

[0113] S45, optimizing the population parameters of the sparrow algorithm according to the clustering index to obtain an optimized population;

[0114] Specifically, the population parameters of the sparrow algorithm are further optimized based on the clustering index. Through this iterative optimization, the clustering effect is improved, ensuring that the clustering results of different load modes and carbon emission modes are more accurate.

[0115] S46. Output an optimal clustering result based on the optimized population; generate an electro-carbon portrait reflecting the characteristics of the set production state based on the optimal clustering result.

[0116] Specifically, based on the optimized population, the optimal clustering results are output to form an electric carbon portrait reflecting the characteristics of different production states. The final clustering results can help the system identify the load and carbon emission patterns under different production states, providing support for further energy management and optimized scheduling.

[0117] In this embodiment, in step S5, the production links with high carbon emission status are analyzed and obtained according to the electro-carbon portrait; and load scheduling and production strategy optimization are performed on the production links with high carbon emission status.

[0118] Specifically, enterprises can analyze the specific production links with high carbon emissions based on the results of electric carbon portraits, and then optimize load scheduling and production strategies.

[0119] Among them, load scheduling optimization: If the profiling results show that a specific electrolyzer exhibits high carbon emissions during the peak period of the electricity market, the company should consider scheduling and optimizing the production tasks during this period. For example, the production tasks during the high carbon emission period can be adjusted to the non-peak period with lower electricity prices and smaller electricity loads, thereby reducing energy consumption and carbon emissions. In addition, companies can also use low electricity price periods to concentrate production to effectively reduce production costs and improve economic benefits.

[0120] Among them, process improvement and equipment upgrade: If the portrait results repeatedly show that a specific equipment or process flow exhibits high carbon emissions at all times, the company should consider upgrading the equipment or optimizing the process for that production link.

[0121] Among them, management decision feedback: the company's management can feedback the causes of high carbon emissions and corresponding optimization strategies to the production department based on the analysis results of the electric carbon portrait, and formulate clear improvement goals and implementation plans based on actual data.

[0122] In summary, the present invention acquires power load data from each production link in real time by setting a data collection strategy; the power load data includes the power consumption and production load status of the electrolytic aluminum equipment; an electrolytic aluminum load model is constructed according to the power load data; the electrolytic aluminum load is divided into set production states through the electrolytic aluminum load model; the set production states include: rated production state, reduced production state, insulation state and cooling state; according to the power load data, in combination with the enterprise carbon quota model and the actual measured carbon emission model, a carbon emission model of the electrolytic aluminum production process is constructed; the carbon emission flow of each production link is quantified through the carbon emission model of the electrolytic aluminum production process, and the electric carbon characteristics under the set production state are obtained; the electric carbon characteristics are input into the SOM network based on the sparrow optimization algorithm for iterative clustering processing, and the optimal clustering result is output; according to the optimal clustering result, an electric carbon portrait reflecting the characteristics of the set production state is generated; according to the electric carbon portrait, the production link with high carbon emission state is analyzed and obtained; load scheduling and production strategy optimization are performed on the production link with high carbon emission state. The present invention combines the load model, free carbon quota and actual carbon emission model of the electrolytic aluminum enterprise, and constructs a multi-dimensional electricity-carbon portrait through carbon emission flow analysis. The portrait integrates the characteristics of electricity use and carbon emission, and divides production states with similar carbon emissions into the same group, helping enterprises to accurately manage carbon emission responsibilities under various load conditions, optimize the use of carbon quotas, and thus effectively improve energy efficiency and promote low-carbon production. In view of the complex and changeable load and carbon emission characteristics of electrolytic aluminum enterprises, the present invention proposes an improved self-organizing map neural network (ISOM) algorithm combined with the sparrow optimization algorithm (SSA). The algorithm overcomes the dependence of the traditional SOM algorithm on manual parameter tuning, and through dynamic parameter adjustment, it better adapts to the real-time load scheduling needs of electrolytic aluminum enterprises under different production conditions, thereby improving the utilization efficiency of carbon quotas and realizing accurate load and carbon emission optimization management.

[0123] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0124] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0125] Example 2

[0126] See also Figure 3 Embodiment 2 of the present invention also provides an electrolytic aluminum electrocarbon image construction device based on an improved self-organizing network, comprising:

[0127] The power load data acquisition module 001 is used to acquire power load data from each production link in real time by setting a data acquisition strategy; the power load data includes the power consumption and production load status of the electrolytic aluminum equipment;

[0128] The electrolytic aluminum load model building module 002 is used to build an electrolytic aluminum load model according to the power load data; through the electrolytic aluminum load model, the electrolytic aluminum load is divided into set production states; the set production states include: rated production state, reduced production state, insulation state and cooling state;

[0129] The electrolytic aluminum production process carbon emission model construction module 003 is used to construct the electrolytic aluminum production process carbon emission model according to the power load data, combined with the enterprise carbon quota model and the actual measurement carbon emission model; the carbon emission flow of each production link is quantified through the electrolytic aluminum production process carbon emission model to obtain the electric carbon characteristics under the set production state;

[0130] The electric carbon image generation module 004 is used to input the electric carbon characteristics into the SOM network based on the sparrow optimization algorithm for iterative clustering processing, and output the optimal clustering result; and generate the electric carbon image reflecting the characteristics of the set production state according to the optimal clustering result;

[0131] The application and optimization module 005 is used to analyze and obtain the production links with high carbon emission status according to the electric carbon portrait; and to perform load scheduling and production strategy optimization on the production links with high carbon emission status.

[0132] In this embodiment, in the electrolytic aluminum load model construction module 002, the expression of the electrolytic aluminum load model is:

[0133]

[0134] P l,min ≤Pl,t ≤P l,max

[0135] Where P l,t I is the power of the electrolytic aluminum load l at time t; l,t is the current intensity of the electrolytic aluminum load l at time t; R l,t is the equivalent resistance; E l,t is the equivalent electromotive force; I max and I min The upper and lower limits of the current allowed to pass into the electrolytic aluminum load l; P l,max and P l,min They are the maximum and minimum values ​​of the electrolytic aluminum load l respectively.

[0136] In this embodiment, in the electrolytic aluminum production process carbon emission model construction module 003, the expression of the enterprise carbon quota model is:

[0137]

[0138] In the formula, E p Total carbon allowances obtained for all electrolytic aluminium loads; E p,lt is the carbon emission quota obtained by the electrolytic aluminum load l at time t; k is the electricity consumption per ton of aluminum; P is the carbon emission quota value obtained per unit of electricity electrolytic aluminum load l; lt is the power of the electrolytic aluminum load l at time t.

[0139] In this embodiment, in the electrolytic aluminum production process carbon emission model construction module 003, the expression of the actual measured carbon emission model is:

[0140] E=E b +E m +E PFC +E e

[0141] Where, E is the total greenhouse gas emissions of the enterprise; E m Carbon emissions caused by the use of energy as raw materials; E PFC is the carbon emissions in the industrial production process; E e Carbon emissions resulting from net purchased electricity consumption.

[0142] In this embodiment, in the electric carbon image generation module 004, the iterative clustering processing submodule includes:

[0143] The parameter initialization submodule 041 is used to initialize the population parameters of the sparrow optimization algorithm;

[0144] The feature data input submodule 042 is used to input the user feature data into the SOM network represented by the sparrow individuals;

[0145] SOM network parameter dynamic adjustment submodule 043, used for dynamically adjusting the learning rate, neighborhood function and weight of the SOM network;

[0146] The SOM network iteration processing submodule 044 is used to input the user feature data into the SOM network after the parameters are dynamically adjusted, perform iterative adjustment until the set number of iterations is reached, and output the SOM clustering results and clustering indicators;

[0147] The population parameter optimization submodule 045 is used to optimize the population parameters of the sparrow algorithm according to the clustering index to obtain an optimized population;

[0148] The electric carbon portrait generation submodule 046 is used to output the optimal clustering result according to the optimized population; and generate the electric carbon portrait reflecting the characteristics of the set production state according to the optimal clustering result.

[0149] It should be noted that the information interaction, execution process and other contents between the modules of the above-mentioned system are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown above in the present application, and will not be repeated here.

[0150] Example 3

[0151] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which a program code for a method for constructing an electrolytic aluminum carbon image based on an improved self-organizing network is stored. The program code includes instructions for executing embodiment 1 or any possible implementation thereof.

[0152] The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0153] Example 4

[0154] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0155] The processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the method for constructing an electrolytic aluminum carbon profile based on an improved self-organizing network according to Example 1 or any possible implementation thereof.

[0156] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor implemented by reading software codes stored in a memory. The memory can be integrated in the processor or can be located outside the processor and exist independently.

[0157] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium, or transmitted from a computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0158] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, and optionally, they can be implemented by a program code executable by a computing system, so that they can be stored in a storage system and executed by the computing system, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0159] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A method for constructing electrolytic aluminum carbon image based on an improved self-organizing network, characterized in that: include: By setting a data collection strategy, real-time power load data is obtained from each production link; the power load data includes the power consumption and production load status of the electrolytic aluminum equipment; Constructing an electrolytic aluminum load model according to the power load data; By means of the electrolytic aluminum load model, the electrolytic aluminum load is divided into set production states; The set production state includes: rated production state, reduced production state, heat preservation state and cooling state; According to the power load data, combined with the enterprise carbon quota model and the actual measured carbon emission model, a carbon emission model for the electrolytic aluminum production process is constructed; the carbon emission flow of each production link is quantified through the carbon emission model for the electrolytic aluminum production process, and the electric carbon characteristics under the set production state are obtained; Inputting the electric carbon characteristics into a SOM network based on a sparrow optimization algorithm for iterative clustering processing, and outputting an optimal clustering result; generating an electric carbon portrait reflecting the characteristics of the set production state according to the optimal clustering result; According to the electric-carbon portrait, the production links with high carbon emission status are analyzed and obtained; and load scheduling and production strategy optimization are performed on the production links with high carbon emission status.

2. The method for constructing electrolytic aluminum carbon image based on improved self-organizing network according to claim 1 is characterized in that: The expression of the electrolytic aluminum load model is: P l,min ≤P l,t ≤P l,max Where P l,t I is the power of the electrolytic aluminum load l at time t; l,t is the current intensity of the electrolytic aluminum load l at time t; R l,t is the equivalent resistance; E l,t is the equivalent electromotive force; I max and I min The upper and lower limits of the current allowed to pass into the electrolytic aluminum load l; P l,max and P l,min They are the maximum and minimum values ​​of the electrolytic aluminum load l respectively.

3. The method for constructing electrolytic aluminum carbon image based on improved self-organizing network according to claim 2 is characterized in that: The expression of the enterprise carbon quota model is: In the formula, E p Total carbon allowances obtained for all electrolytic aluminium loads; E p,lt is the carbon emission quota obtained by the electrolytic aluminum load l at time t; k is the electricity consumption per ton of aluminum; η l p P is the carbon emission quota value obtained per unit of electricity electrolytic aluminum load l; lt is the power of the electrolytic aluminum load l at time t.

4. The method for constructing electrolytic aluminum carbon image based on improved self-organizing network according to claim 3 is characterized in that: The expression of the actual measured carbon emission model is: E=E b +E m +E PFC +E e Where, E is the total greenhouse gas emissions of the enterprise; E m Carbon emissions caused by the use of energy as raw materials; E PFC is the carbon emissions in the industrial production process; E e Carbon emissions resulting from net purchased electricity consumption.

5. The method for constructing electrolytic aluminum carbon image based on improved self-organizing network according to claim 4 is characterized in that: In the process of inputting the electro-carbon characteristics into the SOM network based on the sparrow optimization algorithm for iterative clustering processing, the iterative clustering processing steps are: Initialize the population parameters of the sparrow optimization algorithm; The user characteristic data is input into the SOM network represented by individual sparrows; Dynamically adjust the learning rate, neighborhood function and weight of the SOM network; Input the user feature data into the SOM network after the parameters are dynamically adjusted, perform iterative adjustment until the set number of iterations is reached, and output the SOM clustering results and clustering indicators; Optimizing the population parameters of the sparrow algorithm according to the clustering index to obtain an optimized population; Based on the optimized population, an optimal clustering result is output; based on the optimal clustering result, an electric carbon image reflecting the characteristics of the set production state is generated.

6. An apparatus for constructing an electrolytic aluminum carbon image based on an improved self-organizing network, using the electrolytic aluminum carbon image construction method based on an improved self-organizing network as described in any one of claims 1 to 5, characterized in that: include: The power load data acquisition module is used to acquire power load data from each production link in real time by setting a data acquisition strategy; the power load data includes the power consumption and production load status of the electrolytic aluminum equipment; An electrolytic aluminum load model building module, used to build an electrolytic aluminum load model according to the power load data; By means of the electrolytic aluminum load model, the electrolytic aluminum load is divided into set production states; The set production state includes: rated production state, reduced production state, heat preservation state and cooling state; The module for constructing the carbon emission model of the electrolytic aluminum production process is used to construct the carbon emission model of the electrolytic aluminum production process according to the power load data, in combination with the enterprise carbon quota model and the actual measured carbon emission model; the carbon emission flow of each production link is quantified through the carbon emission model of the electrolytic aluminum production process, and the electric carbon characteristics under the set production state are obtained; An electric carbon image generation module is used to input the electric carbon characteristics into a SOM network based on a sparrow optimization algorithm for iterative clustering processing, and output an optimal clustering result; and generate an electric carbon image reflecting the characteristics of the set production state according to the optimal clustering result; The application and optimization module is used to analyze and obtain the production links with high carbon emission status according to the electric carbon portrait; and to perform load scheduling and production strategy optimization on the production links with high carbon emission status.

7. The device for constructing electrolytic aluminum carbon image based on improved self-organizing network according to claim 6 is characterized in that: In the electrolytic aluminum load model construction module, the expression of the electrolytic aluminum load model is: P l,min ≤P l,t ≤P l,max Where P l,t I is the power of the electrolytic aluminum load l at time t; l,t is the current intensity of the electrolytic aluminum load l at time t; R l,t is the equivalent resistance; E l,t is the equivalent electromotive force; I max and I min The upper and lower limits of the current allowed to pass into the electrolytic aluminum load l; P l,max and P l,min They are the maximum and minimum values ​​of the electrolytic aluminum load l respectively.

8. The device for constructing electrolytic aluminum carbon image based on improved self-organizing network according to claim 7 is characterized in that: In the electrolytic aluminum production process carbon emission model construction module, the expression of the enterprise carbon quota model is: In the formula, E p Total carbon allowances obtained for all electrolytic aluminium loads; E p,lt is the carbon emission quota obtained by the electrolytic aluminum load l at time t; k is the electricity consumption per ton of aluminum; P is the carbon emission quota value obtained per unit of electricity electrolytic aluminum load l; lt is the power of the electrolytic aluminum load l at time t.

9. The electrolytic aluminum electrocarbon image construction device based on improved self-organizing network according to claim 8 is characterized in that: In the electrolytic aluminum production process carbon emission model construction module, the expression of the actual measured carbon emission model is: E=E b +E m +E PFC +E e Where, E is the total greenhouse gas emissions of the enterprise; E m Carbon emissions caused by the use of energy as raw materials; E PFC is the carbon emissions in the industrial production process; E e Carbon emissions resulting from net purchased electricity consumption.

10. The electrolytic aluminum electrocarbon image construction device based on improved self-organizing network according to claim 9 is characterized in that: In the electric carbon image generation module, the iterative clustering processing submodule includes: The parameter initialization submodule is used to initialize the population parameters of the sparrow optimization algorithm; A feature data input submodule is used to input user feature data into the SOM network represented by the individual sparrows; The SOM network parameter dynamic adjustment submodule is used to dynamically adjust the learning rate, neighborhood function and weight of the SOM network; The SOM network iterative processing submodule is used to input the user feature data into the SOM network after the parameters are dynamically adjusted, perform iterative adjustment until the set number of iterations is reached, and output the SOM clustering results and clustering indicators; A population parameter optimization submodule, used to optimize the population parameters of the sparrow algorithm according to the clustering index to obtain an optimized population; The electric carbon portrait generation submodule is used to output the optimal clustering result according to the optimized population; and generate the electric carbon portrait reflecting the characteristics of the set production state according to the optimal clustering result.

Images