Question D Research on the optimization of low-carbon transformation and high-quality development path of China's iron and steel industry
At present, the preliminary code of D question has been written, and the download address is: [2023 Hebei Province Graduate Mathematical Modeling Contest D Preliminary Ideas and Code - 哔哩哔哩] https://b23.tv/g2ATbX5
With the acceleration of my country's industrialization and urbanization process and the continuous upgrading of consumption structure, the rigid growth of energy demand and the increasingly serious problems of resources and environment, energy conservation and emission reduction have become the top priority of the national development strategy. The iron and steel industry is a major energy consumer and carbon emitter. The effect of energy conservation and emission reduction is crucial to the realization of my country's relevant strategic goals and environmental governance, and has become a hot spot that people generally pay attention to. In the context of global low-carbon development, it has become an inevitable choice for China's steel industry to take the road of energy-saving, emission-reducing, low-carbon and green development.
In recent years, my country's iron and steel industry has achieved remarkable results in reducing energy consumption, reducing pollutant emissions, and developing green manufacturing, but there are still many problems. To solve these problems, there is an urgent need for the support of advanced technology, scientific directional guidance, and technical incentives to form an important driving force for the green, low-carbon, high-quality development of the steel industry.
Low-carbon development is of great significance. It will have a profound impact on the steel industry, and even bring about extensive and profound changes, thereby reshaping the regional and even global steel development pattern. Therefore, exploring the influencing factors of the low-carbon development of the world steel industry has important guiding significance for alleviating the emission problems in industrial processes.
The iron and steel industry has the common characteristics of high pollution and high emissions. The factors that affect the CO2 emissions of the iron and steel industry mainly involve the following five aspects:
First, all production activities in the steel industry serve steel production, and crude steel output will directly affect the CO2 emissions of the steel industry.
Second, the urbanization process of countries around the world needs the support of the iron and steel industry. Urbanization means that the demand for construction, automobiles, home appliances and other manufacturing industries and infrastructure construction will increase. Steel is the most important material in construction and one of the main materials in automobile production. It is also widely used in home appliances or other equipment manufacturing industries. Infrastructure investment and construction will drive the demand for steel, thereby affecting the overall CO2 emissions of the steel industry. quantity.
Third, the continuous deepening of the globalization trend has provided the possibility for the further expansion of the trade opening of the iron and steel industry in all countries in the world. The expansion of trade openness in the iron and steel industry has enabled the production factors of the world's iron and steel industry to be fully utilized, optimized resource allocation, improved the structure of the iron and steel industry in various countries, improved the level of green production technology, and reduced the overall CO2 emissions of the world's iron and steel industry.
Fourth, the Environmental Kuznets Curve assumes an "inverted U"-shaped relationship between economic growth and environmental pollution. When a country's economic development is at a low level, the increase in per capita GDP will lead to an increase in environmental pollution. When per capita GDP increases to a certain level, the level of environmental pollution will reach its peak. With the further increase of per capita GDP, environmental pollution will be improved.
Fifth, financial development is conducive to the iron and steel industry's access to funds for energy conservation and emission reduction, and financial development has a negative impact on the CO2 emissions of the iron and steel industry to a certain extent.
Now we have collected the CO2 emissions of China, India, Japan, the United States, Russia and the world's steel industry and the world's crude steel production from 2000 to 2019, and the CO2 emissions per ton of crude steel in China, India, Japan, the United States and Russia from 2015 to 2019 The amount of heat consumed by blast furnace gas and the proportion of electric arc furnace steelmaking. See Appendix 1 "Fill in the Blanks by Data Inference".
• The CO2 emission per unit of crude steel is an important indicator to evaluate the low-carbon development of a country's steel industry. The smaller the value, the smaller the CO2 emission per unit of crude steel.
• Blast furnace gas is a by-product of blast furnace ironmaking. It is a combustible gas and can be used to heat hot-rolled steel ingots and preheat ladles. The main components of blast furnace gas are carbon monoxide, carbon dioxide, nitrogen, hydrogen and methane. The heat consumed by blast furnace gas includes energy transformation consumption, industry self-use consumption, loss and statistical error, which can be used as an indicator to measure a country's low-carbon development.
• Compared with the traditional "blast furnace-converter" long-process steelmaking process, the CO2 emissions produced by electric arc furnace steelmaking are only 1/3 of the former. Electric arc furnace steelmaking can effectively reduce CO2 emissions in the world's steel industry, which is a breakthrough The reliance on long-process steelmaking is an important means to adapt to the new pattern of low-carbon development of the world's steel industry.
This question aims to clarify the trend of CO2 emissions in the iron and steel industry since 1990 and the contribution rate of my country in the low-carbon transformation of the world iron and steel industry by solving the following four problems, and quantitatively characterize the crude steel production (PROUD, unit: thousand tons) , urban population (URBAN), steel industry trade openness (COPEN) calculation related indicators (unit: thousand tons), per capita GDP (RGDP, 2010 constant price USD) and financial development index (FD, value 0 to 1, The larger the value, the better the degree of financial development) on the CO2 emissions of the iron and steel industry, recommending the optimal path for the high-quality development of China's iron and steel industry.
1. Determine the quantitative relationship between the CO2 emissions of the world's iron and steel industry and the CO2 emissions of China, India, Japan, the United States, and Russia's iron and steel industries, fill in the missing data marked yellow in the attached file "Data Inference Fill in the Blanks", and build a mathematical model to illustrate the The basis for filling in the data.
Multiple linear regression methods can be used
Multiple linear regression is a commonly used predictive modeling method in statistics and machine learning. It is built on the basis of simple linear regression, allowing the influence of multiple independent variables on the dependent variable to be considered simultaneously. The following is a summary of the main points of multiple linear regression:
Model representation: The multiple linear regression model is expressed as a mathematical expression: Y = β₀ + β₁X₁ + β₂X₂ + ... + βᵣXᵣ + ε where, Y is the dependent variable (the target variable to be predicted), X₁, X₂, ... , Xᵣ is the independent variable (features that affect the dependent variable), β₀, β₁, β₂, ..., βᵣ are the regression coefficients, indicating the degree of influence of the independent variable on the dependent variable, ε is the error term, representing other factors that have not been considered Effects on the dependent variable and uncertainty in the model.
Regression Coefficient: The regression coefficient βᵢ measures the influence of each independent variable Xᵢ on the dependent variable \(Y\). The sign of the coefficient indicates the direction of the effect (positive or negative correlation), while the magnitude of the coefficient indicates the strength of the effect. The goal of a regression model is to find the optimal regression coefficients that minimize the error between the predicted value and the true value (usually achieved using the least squares method).
Model assumptions: Multiple linear regression relies on a number of assumptions, including linear relationships, normal distribution of error terms, independence between independent variables, and homoscedasticity of error terms. Violations of these assumptions may result in inaccurate models.
Goodness of fit: Goodness of fit (R-squared) is used to evaluate how well the model fits the data. It represents the proportion of the variation in the dependent variable that can be explained by the independent variable. The value range of R-squared is between 0 and 1. The closer to 1, the better the model fit, and the closer to 0, the poorer the model fit.
Model evaluation: In order to evaluate the performance of the model, various metrics can be used such as Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) wait.
Feature Engineering: When building a multiple linear regression model, choosing the right independent variables is critical to model performance. Feature engineering involves selecting meaningful features, handling missing values, handling outliers, performing data transformations, and more.
Forecasting and inference: Multiple linear regression models can be used to predict future outcomes, as well as to infer relationships between variables. Forecasting is estimating the value of the dependent variable by entering new values of the independent variable, while inference is performing a significance test on the regression coefficients to determine whether the independent variable has a significant effect on the dependent variable.
Overall, Multiple Linear Regression is a powerful modeling technique that can be used to explore relationships between variables, and to perform predictive and inferential analyses. However, building a reasonable model requires attention to the quality of the data, the selection of feature engineering, and the verification and correction of model assumptions.
% 输入数据
year = [2000:2019]'; % 年份
china_CO2 = [2.52 2.68 2.74 3.27 3.76 4.89 5.64 6.66 7.19 8.43 9.45 10.52 11.01 11.86 12.40 11.86 11.65 11.35 12.62 12.89]'; % 中国钢铁工业CO2排放量
india_CO2 = [0.70 0.67 0.82 0.80 0.82 0.86 1.00 1.11 1.14 1.39 1.56 1.83 1.85 2.00 2.25 2.31 2.45 2.61 2.93 2.92]'; % 印度钢铁工业CO2排放量
japan_CO2 = [0.78 0.76 0.78 0.81 0.81 0.85 0.86 0.89 0.83 0.76 0.88 0.90 0.89 0.89 0.87 0.83 0.81 0.83 0.81 0.79]'; % 日本钢铁工业CO2排放量
usa_CO2 = [0.91 0.76 0.53 0.58 0.68 0.50 0.54 0.51 0.52 0.33 0.49 0.51 0.48 0.49 0.49 0.47 0.47 0.51 0.48 0.48]'; % 美国钢铁工业CO2排放量
russia_CO2 = [0.75 0.73 0.75 0.71 0.73 0.83 0.71 0.92 0.86 0.85 0.91 1.01 1.38 1.34 1.44 1.46 1.48 1.47 1.54 1.70]'; % 俄罗斯钢铁工业CO2排放量
world_CO2 = [8.75 8.74 8.56 9.10 10.10 10.83 11.65 12.84 13.57 13.93 15.57 17.40 17.94 18.76 19.86 19.03 18.75 18.84 20.39 21.12]'; % 世界钢铁工业CO2排放量
% 构建设计矩阵X
X = [china_CO2, india_CO2, japan_CO2, usa_CO2, russia_CO2];
% 使用多元线性回归模型拟合数据
mdl = fitlm(X, world_CO2);
2. Combined with the previous research, considering the crude steel production in the world, crude steel production in China, India, Japan, the United States, Russia, and the one-to-one data relationship between crude steel production and steel industry CO2 emissions, construct a mathematical model to determine China’s , India, Japan, the United States and Russia's crude steel production from 2000 to 2019, and verify the accuracy of the solution results by ourselves.
3. According to the attached document "Influencing Factor Data", determine the calculation model of CO2 emissions from the steel industry in China, India, Japan, the United States and Russia, and empirically analyze the sensitivity and robustness of the calculation model.
The visualization is as follows:
4. Combining the research results of the above three issues, construct an optimization model for the lowest CO2 emissions of China's iron and steel industry, and formulate an optimization model for the low-carbon transformation and high-quality development of China's iron and steel industry based on the model calculation result data or the previous research result data path.