Overview of the Development of Coking Technology in China and the Current Status of Coking Technology Research in Our Institute

Thread Content

An Overview of the Development of Coking Technology in China and the Current Status of Coking Technology Research in Our Institute Author: Shui Hengfu (Excerpted from Coal Coking Network) 1. An Overview of the Development of Coking Technology in China In 2006, China’s coking industry experienced steady and rapid development; the annual production of coke reached 290 million tons, with 14.5 million tons exported. The apparent domestic consumption of coke was 270 million tons, making China the country with the highest levels of both coke production and consumption in the world. The increase in coking scale has also promoted the development of coking technology in our country. In recent years, the main developments in coking technology in our country can be summarized as follows. 1.1 The trend is toward larger coke ovens; larger coke ovens are an inevitable requirement for the sustainable development of the coking industry. Large coke ovens not only improve labor productivity and reduce environmental pollution during the coking process, but also enhance the quality of coke. The \"Access Requirements for the Coking Industry\" issued in China in 2004 stipulated that coke ovens with a capacity of less than 4.3 meters must be phased out, further promoting the trend toward larger coke ovens in the country. The successful commissioning of the 7.63m coke ovens at Shandong Yankuang International Coking Company, Taiyuan Iron and Steel Co.’s coking plant, and Maanshan Iron and Steel Co.’s coal coking company indicates that China now possesses the capability to produce ultra-large-scale coke ovens, thereby improving the technical level of the country’s coking industry. At the end of the 1980s, European coking industry experts proposed the idea of changing traditional multi-chamber coking ovens, and conducted tests using a large-scale coking reactor (referred to as JCR) for a scheme that combined coal preheating with direct coke dry quenching, achieving good results. The quality of coke sees an increase in CSR of 4–10 percentage points, while CRI improves by 2–4 percentage points. The range of coal types has been expanded, allowing for greater use of coals with high expansion, low volatiles, and weak caking properties. It offers significant energy-saving effects, with an 8% reduction in energy consumption. Reduce pollutant emissions by 50%. Based on the test results of the JCR process, an industrially viable \"single-chamber coking system (abbreviated as SCS)\\" has been developed. The recommended basic parameters for SCS to increase the output per unit furnace volume, save on investment, and significantly enhance the structural strength of the furnace (for implementing the coal preheating coking process) are: H×L×W = 9.5m×19m×(450～610)mm. 1.2 Coal moisture control technology (abbreviated as CMC) refers to the process of pre-dehydrating and drying wet coal to a moisture content of less than 6% outside the furnace before feeding it into the furnace for coking; this is known as the dried-coal coking process. The coal moisture-adjusted coking process was developed on the basis of the dry coal coking process; its core principle is to keep the moisture content of the coal fed into the furnace within the range of 5% to 6%, regardless of the moisture level of the raw coal. The basic principle of coal drying and coal moisture regulation is a coking technology that uses external heat energy to dry and dehydrate coking coal outside the coking furnace, in order to reduce the moisture content of the coal fed into the furnace or to adjust its moisture level. This approach helps to control the heat required for coking, improve the operation of the coking furnace, enhance the quality of coke, and increase the use of weakly bonded coal. CMC technology was first introduced and used in Japan. The first-generation CMC at Nippon Steel in Japan uses a hot kerosene drying method, in which the sensible heat of the gas from the coke oven rising pipes and the waste heat from the coke oven flue gases are utilized to heat kerosene, which then indirectly dries the coal material in a multi-tube rotary dryer. The second-generation CMC is dried using steam. Using the low-pressure steam generated from the dry quenching of coke, or other low-pressure steam from the plant, as a heat source, the coal is indirectly dried with steam in a multi-tube rotary dryer. The first set of this CMC was built and put into operation at the Kunisaki Plant of Nippon Steel in March 1991. In October 1996, the third-generation fluidized-bed CMC unit was developed and put into operation at the Muroran Plant of Hokkai Steel Co., Ltd.; this unit was constructed in conjunction with the No. 5 coke oven at the Muroran Plant, and its coal processing capacity is 120 t/h. The annual output of coke is nearly 700,000 tons. China’s first CMC unit was put into operation at Chongqing Iron and Steel (Group) in 1996. At the beginning of this century, due to shortages of coking coal resources and energy, the process of drying coal for coking gained attention and was promoted for use in China. The coking plants of Liaoning Benxi Iron and Steel Company, Henan Pingdingshan Tianhong Coking Company, Shaoxing Iron and Steel Company, Xiangtan Iron and Steel Company, etc., have successively adopted the dried coal coking process in their coal preprocessing procedures, and have improved upon the existing processes to develop a coal moisture-adjustment process. 1.3 Ramming coking technology: Compared with top-charging coke ovens, ramming coking allows for more rational utilization and expansion of coking coal resources, improves the quality of coke, and enhances labor productivity. M40 can increase by 1%–6%, M10 can see an improvement of 2%–4%, CSR increases by 1%–6%, and production capacity rises by 10%. Due to the advantages of the rammed coking technology, and especially given the shortage of primary coking coal in our country, its development has been rapid, and it has become the preferred method for coking production in some regions. The 55-chamber, 5.5m ramming coke ovens of Yunnan Yunwei Group Co., Ltd. came online at the end of 2006; the coal loading capacity per chamber exceeds 35 tons, which is close to the world’s advanced levels, and this has laid a foundation for the development of ramming coke-making technology in China. This ramming coke oven can even be used for coking with anthracite, thereby greatly expanding the sources of coal available for coking. This aligns with the development direction of coking technology in China, and it holds broad prospects for further development. 1.4 Briquette coking technology: Briquette coking technology can improve the quality of coke and reduce the amount of highly cohesive coal used. The main reasons for this are as follows: (1) The coal particles within briquettes are in close contact with each other, which facilitates the combination of bonding and non-bonding components during the coking process, thereby improving the coking properties of the coal. (2) When briquetted coal is mixed with pulverized coal for coking, during the softening and melting stage, the expansion of the briquetted coal’s volume leads to closer contact between the coal particles, resulting in coke with a firm structure. (3) In the coal charged to the furnace along with briquetted coal, due to the high density of the briquettes, their thermal conductivity is better than that of pulverized coal; as a result, the heating rate is faster, allowing the softening temperature to be reached earlier, which extends the plastic temperature range of the pulverized coal. (4) Coking coal feedstock with briquetted coal has a high bulk density, resulting in less shrinkage of the semi-coke during the coking process, and thus fewer cracks in the coke. (5) A certain amount of binder was added during the shaping of the coal to be charged into the furnace, which has a certain effect on modifying the coking coal material. Baosteel is a representative example of coking processes using blended coal in China. Baosteel’s Phase I project adopted the coking process using Nippon Steel’s blended coal. Phase 3 introduced Nippon Steel’s streamlined coal blending coking process. The implementation of this technology has played a positive role in improving the quality of metallurgical coke produced by Baosteel and alleviating the shortage of coking coal in the East China region. Coking plants such as Wugang Coking Plant and Shuicheng Coking Plant use briquette coking technology with tar residue added, which enables effective treatment and utilization of tar residue while improving the quality of coke. 1.5 Dry quenching technology (abbreviated as CDQ) boasts outstanding advantages in terms of environmental protection, energy savings, and improvement in coke quality; as a result, it has developed rapidly in China. Currently, 44 dry quenching units are in operation in China, with a coke processing capacity of over 30 million tons. The sensible heat of the red coke produced is responsible for 35% to 40% of the energy consumption in coke ovens, whereas dry quenching allows for the recovery of 80% of the sensible heat from the red coke. Operational experience with Baosteel’s dry quenching system shows that it can reduce energy consumption by an average of 50–60 kg of standard coal per ton of coke. The production practices of dry quenching coke at Magang show that the M40 value of coke can increase by 3%, the M10 value can decrease by more than 1%, the CSR value can increase by more than 3%, and the CRI value can improve by more than 1%. Since the dry quenching system operates in a closed environment, it causes little environmental pollution, helps to reduce greenhouse gas emissions, and also saves water; its economic and social benefits are significant, making it worthy of further promotion. As China gains mastery over the manufacturing capabilities for key equipment used in dry quenching systems and accumulates production experience, coupled with the increasing demands for environmental protection in coking processes, dry quenching technology is bound to be widely adopted and developed in China. 1.6 Production of methanol from coke oven gas Methanol is an important chemical raw material as well as a key energy source for the new generation. Methanol can be further used to produce dimethyl ether (DME), as well as gasoline (MTG). Methanol is the parent compound of carbon-1 chemistry, and thus has a wide range of applications. Coke oven gas is a hydrogen-rich gas, containing 55%–60% H2, 5%–8% CO, and 23%–27% CH4, making it an ideal gas as a raw material for the chemical industry. The production of methanol from coke oven gas is in line with **’s industrial, energy, and environmental protection policies, and has become a new source of economic growth for coking enterprises. China’s first methanol production project using coke oven gas, with an annual capacity of 80,000 tons, designed by the Second Design Institute of the Chemical Industry, was successfully put into operation in December 2004 at the Coking and Gas Production Company in Qujing City, Yunnan. At present, the capacity for producing methanol from coke oven gas, either already in operation or under construction, has reached over 1 million tons per year, indicating good prospects for further development. In particular, when combined with coal-to-gas production, it enables more rational utilization of carbon and hydrogen resources, allows for maximum use of coke oven gas, increases methanol production, and reduces the production costs of methanol. 1.7 Centralized processing of coal tar: With the rapid development of China’s coking industry and the enlargement of coke ovens, centralized processing of coal tar has become an emerging trend. Due to the shortage of coal tar resources and the continuous rise in its prices, small coal tar processing enterprises have been gradually phased out, replaced by a new batch of coal tar processing plants with capacities of 200,000–300,000 tons per year. Some old large-scale coking plants are also continuously expanding to increase the scale of coal tar processing. At present, coal tar processing units with a single processing capacity of 300,000 tons per year have been built and put into operation in Shanxi and other regions. The expansion of coal tar processing scale has also enabled some new technologies that combine atmospheric and vacuum distillation from the refining industry to be applied in coal tar processing plants, significantly improving the technology used for processing coal tar. 2 Current Research Status of Coal Coking Technology in Our University 2.1 Basic Theoretical Research on Coal Cohesiveness Cohesiveness is an important technological property of coal used for coking; understanding the mechanism behind the formation of coal cohesion theoretically is of great significance for the rational processing and utilization of coal. Currently, the colloid theory and intermediate phase theory are widely applied, but the key components that affect coal bonding properties have not yet been identified. In addition to the complexity of coal itself, another reason is the lack of a solvent that exhibits excellent solubility for coal. Since Lino and others discovered that the mixed solvent of carbon disulfide and N-methyl-2-pyrrolidone (CS2–NMP) can achieve high extraction rates of 50% to 80% for certain bitumens at normal temperature and pressure, this mixed solvent has been widely used in the study of coal structure. The agglomeration structure of coal has a significant impact on its physical properties such as solubility, thermal expansion coefficient, electrical conductivity, and specific heat. The extraction rate of coal samples is related to their cohesion, and this extraction rate is closely connected to the agglomeration structure of the coal. It can be inferred that the association structure of coal is closely related to its bonding properties, yet there are no existing literature reports on this aspect. The cohesion of coal samples is closely related to their extraction rate; plotting the extraction rate using a mixed solvent along with the cohesion index yields the graph shown in Figure 1. http://www.meijiaohua.com/WebUpLoadFile/ArticlePhoto/2007101531508345.JPG Figure 1 Relationship between coal sample extraction rate and bonding index GR.I (△—NMP content in the mixed solvent < 50%) ； □—(In mixed solvents with an NMP content >50%) As can be seen from Figure 1, when the three coal samples are extracted using CS2-NMP mixed solvents in different proportions, and the NMP content in the mixed solvent is <50%, the bonding index GR.I of the coal samples decreases gradually as the extraction rate increases; whereas when the NMP content >50%, the bonding index GR.I of the coal samples increases gradually as the extraction rate decreases. It can be seen that the bonding index of the coal samples all shows a decreasing trend as the extraction efficiency using CS2-NMP mixed solvents in different proportions increases. This shows that the caking property of coal is closely related to its extraction rate. Furthermore, for the coking coal and bituminous coal used in our experiments, their bonding indices differ slightly at the same extraction rate. As can be seen from Figure 1, at the same extraction rate, the bonding index of the raffinate obtained using mixed solvents with different proportions in which the content of coal sample in NMP is >50% is higher than that obtained using mixed solvents with different proportions in which the content of coal sample in NMP is <50%. It shows that although coal samples have the same extraction rate in mixtures of solvents at different proportions, their residues and extracts still differ in composition after extraction, which leads to variations in the bonding index GR.I of the coal sample residues. For lean coal, the extraction rate is relatively low. To this end, the PI of rich coal (soluble in mixed solvents, insoluble in pyridine) was added to lean coal to examine its effect on the extraction rate and cohesion of the lean coal. The effects of the soluble components on the extraction rate and cohesion of weakly caking and lean coals are shown in Table 1. Table 1 Changes in the extraction rate and cohesion of lean coal. Coal sample | Extraction rate | Cohesion index: Lean coal – 5.1, 17; Lean coal + rich coal PI (0.1 g/g coal) – 7.9, 56. As can be seen from Table 1, the addition of the soluble component, the PI of rich coal (0.1 g/g coal), to lean coal results in a slight increase in its extraction rate. We believe that the extraction process takes place at room temperature, making chemical reactions almost impossible. During the dissolution in the CS2-NMP mixed solvent, the PI molecules of lean coal may reassociate with those of rich coal, weakening the interactions within the aromatic ring systems in the coal. This disrupts the π‑π bonds and van der Waals forces between coal molecules, resulting in a slight increase in the extraction rate of lean coal; moreover, the PI molecules of rich coal have a solubilizing effect on lean coal. However, after adding the soluble component, PI of fat coal, to lean coal, its bonding index increased significantly. It can be seen that the PI of fat coal has a very significant impact on the bonding properties of lean coal, acting as a binder. This may be due to the fact that when the coal sample is heated and carbonized, not only do the soluble components of lean coal break free from their non-covalent bonds and escape, but it is also possible that the small molecules of PI from rich coal recombine with those from lean coal. The other small molecules resulting from the breakdown of covalent bonds, together with these PI molecules from rich coal, form a plastic flow phase that gives rise to colloids, thereby significantly increasing the bonding index of the coal sample. 2.2 Study on the modification of coking coal by hydrothermal treatment. The association structure of coal affects its solubility and thermal properties in organic solvents. The agglomeration structure of coal also has a significant impact on its reactivity and transformation. Pre-treating coal in order to alter its association structure and thereby change its properties is a highly meaningful task. Water treatment is a very important pretreatment method, as it can alter the reactivity and structure of coal. Graff and Brands believe that treating Llinois №6 coal with water at 50 atmospheres of pressure and temperatures between 340–500°C can increase its extraction rate from 17% to 30%, but after exposing the pretreated coal to air for a few minutes, its extraction rate becomes almost the same as that of the untreated coal. Bienkowski et al. found that the liquefaction rate of coal treated with water at 200°C increased from 27.3% to 38.4%. Lino et al. found that the extraction rate of coal after treatment with water at 600°C increased significantly. As a result, there has been a large amount of research on the extraction and liquefaction mechanisms of coal after water treatment. Theories regarding the breaking of weak bonds in coal after water treatment, such as the theory of ether bond breaking, the theory of hydrogen bond breaking, the theory of mineral transfer during water treatment, and the theory of reduction of oxygen-containing groups. To this end, we conducted hydrothermal treatment experiments on bituminous coal and lean coal, and found that the extraction rate and G value of the coal after treatment changed significantly, as shown in Table 2. Table 2 G values of gas coal and lean coal, as well as the extraction rates (wt%) using the CS2-NMP mixed solvent at room temperature. Item: Gas coal; Lean coal. G value; Extraction rate. G value; Extraction rate of raw coal: 74.80; 27.32; 18.51; 6.53. Water treatment 1: 76.80; 34.21; 17.73; 6.24. Water treatment 2: 60.51; 30.48; 21.62; 21.29. Water treatment 3: 50.43; 27.02; 17.52; 9.22. Water treatment 4: 56.82; 36.28; 16.42; 5.38. It can be seen that water treatment of coal prior to coking reduces the reactivity of the coke and increases its strength after reaction. The main reason for this is likely that water treatment removes some of the metallic minerals present in the coal, while also causing the breakage of alkyl side chains and more reactive bonds within the coal molecules, thereby reducing the reactivity of the coke. 2.3 Development of an expert system for coking coal blending and prediction of coke quality: By leveraging computer science and the knowledge and experience accumulated by experts in coal blending, a system is established to predict coke quality, manage production, ensure quality control, and achieve low-cost coal blending suitable for the characteristics of domestic coal. In practice, the core elements such as the evaluation of the properties of individual coal types, coal resource databases, expert system knowledge bases, the utilization of weakly caking coal, and optimization control algorithms have been further improved. 2.3.1 General approach First, establish the basic information for a database of coal resources used in enterprise coking, including important coal quality data as well as related information ； Secondly, by leveraging knowledge from information science, computer science, and control theory, and integrating the practical experience of renowned coal blending experts, the basic framework and software design platform for a coking coal blending expert system are designed and proposed ； Third, within the structure of the aforementioned system, historical data on coal used in enterprise production is analyzed, or data on individual coal sources that could be utilized is analyzed, in order to obtain relevant information for the coal resource database ； Fourth, conduct single-coal coking tests on the coal currently in use or those that may be used in the future, in order to obtain information regarding the coking properties of each type of coal and its contribution to coke quality ； Fifth, determine the constraints for the use of each individual coal type based on the expert knowledge base ； Sixth, linear programming, nonlinear programming, or simulated evolutionary algorithms are used to determine the optimal coal blending ratio. The objective function for optimization is to minimize the coal blending cost while ensuring the quality of coke ； Seventh, the predicted value of coke quality under the optimal ratio is calculated using a coke quality prediction model, and after verification by a coal blending expert system or experts, the optimal coal blending ratio for production is determined ； Eighth, computer-based automatic control is used for operating the coal blending tank, ensuring a precision in controlling the coal blending ratio of over 97% ； Finally, the self-learning of the performance-based model and the adjustment of the control model are carried out. 2.3.2 Main technical contents (1) Coking coal resource system. From an application perspective, the coking coal resource utilization database system is relatively independent, comprising sub-systems such as an overview of global coal resources, an overview of coal resources in China, and an overview of coal resources used by enterprises or companies. (2) Principles and design of the coking coal blending expert system. According to the current framework of coking production, its functions are structured such that the software design is divided into three main modules: the information management module, the data model module, and the ratio generation module. These are described as follows: Information Management Module: For enterprises that use large databases, general statistical and management functions are handled by this system, while the expert system is responsible only for data collection and transmission. Companies without a database design their own database systems, which coexist with production data and historical data. Data model section: It provides relevant data model modules, as well as several knowledge base data sources, to support the enterprise’s quality engineer (6 Sigma) projects. The proportioning section: Achieving the prediction of coke quality based on proportions and optimizing the proportioning scheme are the core of technological innovation. Its function is to meet certain coke quality requirements, to facilitate the derivation of optimized composition schemes based on various constraints, or to provide quick coke quality predictions once a specific composition is given. (3) Database of properties of individual coking coals and its evaluation system. Based on the conventional use of database functions and structures, and taking into account the specific characteristics of the coking industry, a system for characterizing and evaluating individual coal types was designed and improved. This system introduces new concepts such as coal quality variability, authenticity, usage constraints, and cost-performance ratio, providing highly valuable technical support for the rational planning of coal resources and the optimization of coal blending. The coal quality database includes historical data from a certain period as well as all current production data, enabling functions such as input, output, querying, browsing, adding, and modifying of production data (for individual coals, blended coals, coke, coke oven operations, etc.), statistical analysis, and performance identification. Comprehensive analysis can also be conducted on various charts, as well as the trends and fluctuations of different indicators. Regarding the evaluation of single coal types, most researchers in previous studies have provided only qualitative descriptions. This technique establishes a quantitative evaluation system and usage constraints based on historical and experimental data, thereby determining the coking properties and cost-performance ratio of the coal quality. (4) General platform for quality prediction and control models. A general platform for designing new methods of coke quality prediction has been developed, suitable for different types of coal blending configurations and coke quality requirements. The research on the influencing factors and control methods of the thermal properties of coke is comprehensive, in-depth, and meticulous; moreover, the practical conditions of enterprises are taken into account during this research, thereby continuously enriching its content. The prediction models for coke quality include forecasts of coke ash content, sulfur content, M40, M10, CSR, and CRI. When developing these various prediction models, existing knowledge, expert experience, and research findings are utilized; moreover, depending on the way in which different types of data contribute to the prediction of coke quality, methods such as weighted summation, adjusted weighting, and the use of penalty/reward coefficients are employed to refine the prediction results. Given the complex factors that influence indicators such as M40, M10, CSR, and CRI in coke, and considering that the thermal properties of coke are widely studied, the prediction models consist of two approaches: one based on data from a single coal type and another based on data from blended coals. The data from a single coal type is used to derive information about blended coals, which in turn is used to predict coke properties. 2.4 Development of the automatic heating control system for coke ovens. Coke ovens are the most complex furnaces in the metallurgical industry. The heating process in these ovens is a thermal process that involves intermittent operation in individual combustion chambers, continuous operation across the entire furnace, and is affected by various factors; it represents a typical complex system with high inertia, nonlinear behavior, and fast time variations. It was first developed and applied by Nippon Steel Corporation at Coke Oven No. 5 in Fukuyama in 1973. There are currently over a dozen process flows and control methods; representative processes include Japan’s CCS system, the U.S.’s COHC system, Germany’s CODECO and ABR systems, France’s CRAPO, Belgium’s Center for Metallurgical Research CRM, and the Netherlands’ CETCO process. There are also various forms of control methods; one of them involves setting the gas flow rate or heat supply based on the parameters of the coal used in firing and the production parameters ； The second method is to maintain stability in the flue temperature; by adjusting the gas flow based on the deviation between the actual flue temperature and the set flue temperature, the flue temperature is kept stable ； The third control method is essentially a combination of the first two, that is, a combination of feedforward and feedback regulation. Currently, this technology is in normal operation in countries such as Japan, Germany, the United States, the Netherlands, France, and Belgium. The block diagram of the cascade control system for coke oven heating optimization developed by Anhui University of Technology is shown in Figure 2. Figure 2 Block diagram of the optimized cascade control system. The first approach is a scheme to stabilize the coking time: it involves optimized cascade control based on a combination of two feedforward elements, two feedback elements, and one monitoring element. The two feedforward controls are heat supply feedforward and flue suction feedforward control ； The two types of feedback are furnace temperature feedback and target temperature feedback ； One type of monitoring is the monitoring of the oxygen content in the smoke ducts (or α feedback) ； The furnace temperature is controlled using cascade control, while the suction force is controlled via a set-point following control scheme. The second control scheme is used when the coking time changes: it involves a combination of optimized cascade control and an expert system, and this scheme is activated only when the coking time changes, with adjustments being made step by step depending on the extent of those changes. It features simple operation and stability and reliability; once the specified coking time and furnace temperature are reached, it enters the control mode for a stable coking time. The key to achieving optimized cascade control in coke ovens is to establish a reasonable and accurate process mathematical model. The methods for establishing mathematical models can be divided into theoretical modeling and experimental modeling. The former establishes mathematical expressions that describe the dynamic characteristics of an object through analysis based on its internal mechanical processes. Due to the complexity of the coke oven heating system, some simplifications and assumptions must be made when modeling in this way, which undoubtedly leads to errors and reduces the validity and accuracy of the mathematical model. Experimental modeling involves testing an actual production process or object to obtain input and output data, which is then analyzed and processed to establish input-output models; it is a very simple and effective method for complex systems. The development of this model was based on the coke oven heat balance theory and the principle of double-layer flat walls for heat transfer in coke ovens; calculation formulas for the heat supply were derived through analysis and comparison. A feedforward heat supply model (experimental modeling) was proposed, which is grounded in the coke oven heat balance theory and utilizes a large amount of data collected during the production process for processing and calculation ； Based on the heat transfer principles of coke ovens, and utilizing a large amount of data obtained from online measurements, relevant models for the target channel temperature, the center temperature of the coke cake, and the channel temperature were developed, as well as a coking index model to determine when coking is complete. The collection of data for modeling optimized cascade control in coke ovens is divided into two categories: one is the collection of historical data over several years ； Another category is collecting relevant data on-site. Practical operation on more than 50 coke ovens has proven that the automatic heating control system developed by Anhui University of Technology operates stably and reliably, offers significant energy-saving benefits by enabling approximately 2% savings in gas usage, and also improves the quality of coke. This post was last edited by ryn on 2009-3-10 17:18]