Operation method of integrated gasification combined cycle using data labeling automation technology

Abstract

The present invention provides an operation method of an integrated gasification combined cycle, the method comprising the steps of: (a) securing basic data obtained from one or more raw materials and additives, one or more sensor means installed in the integrated gasification combined cycle, and intrinsic values of the raw materials and additives; (b) securing process data obtained from the sensor means; (c) securing output data obtained through the integrated gasification combined cycle; (d) collecting the basic data, the process data, and the output data to label the data with a machine learning model; and (e) deriving optimal operating conditions on the basis of the labeled data.

Description

Operation method of coal gasification combined cycle power generation using data labeling automation technology

[0001] The present invention relates to a method for maximizing the operating efficiency of a coal gasification combined cycle power generation system using data labeling automation technology, and more specifically, to a method for proposing optimal operating conditions by applying basic data regarding raw coal and additives, process data from various sensors generated during plant operation, and output data regarding quality and production information regarding the final energy source to data labeling automation technology.

[0002] Although coal is one of the raw materials with guaranteed supply stability due to its abundant and extensive global reserves, the volatility of coal prices has increased as the global supply and demand structure undergoes real-time changes driven by factors such as rising demand for thermal coal in countries like China and India, and export restrictions imposed by major coal-producing nations. Furthermore, demand from developed countries has begun to rise as the value of clean coal utilization technologies, such as the Integrated Gasification Combined Cycle (IGCC), which emits lower levels of pollutants like carbon dioxide and sulfur dioxide compared to conventional coal power generation, increases. Consequently, as coal supply volumes and costs fluctuate sensitively depending on international affairs and natural disasters, it is necessary to diversify coal sources and secure process technologies capable of maintaining a certain level of gasification efficiency to ensure the stable operation of IGCC.

[0003] Korean Registered Patent Publication No. 10-2200408 relates to an apparatus and method for predicting operational abnormalities of a coal gasification plant using a self-learning model, comprising: a performance prediction unit that derives a performance prediction value by predicting the performance of the equipment at a prediction time point after a predetermined time from the reference time point based on operating data at the reference time point through an artificial neural network; an operation guide unit that derives a performance calculation value by calculating the performance of the equipment based on operating data at the prediction time point through a physical model; and a self-learning unit that determines a value having a dominant error among the performance prediction value and the performance calculation value by comparing the performance prediction value and the performance calculation value with respect to a performance statistical value, which is a statistical value of the equipment's performance corresponding to past operating data that is similar to the operating data at the prediction time point by a preset value or more, and modifies either the artificial neural network or the physical model according to the determined value.

[0004] Korean Registered Patent Publication No. 10-2200407 relates to an operation guide system for a coal gasification plant and an apparatus for the same, comprising: a fuel determination unit that determines the gasification suitability of a fuel selected by a user when a fuel to be analyzed is selected according to user input before the plant starts; a performance analysis unit that performs a performance analysis of the plant, including the performance of a gasifier and the performance of a syngas cooler, during plant operation; and an operation guide unit that provides an operation guide indicating a control value for plant operation based on the performance analysis; wherein the fuel determination unit distinguishes and outputs whether the fuel to be analyzed is coal usable without flux or blending, coal usable when flux is injected, coal requiring blending, or coal unsuitable for gasification based on the result of determining the gasification suitability.

[0005] Japanese Registered Patent Publication No. 6080567 relates to a method for controlling the operation of a coal gasification combined cycle power plant and a coal gasification combined cycle power plant. The method provides a method for controlling the operation of a coal gasification combined cycle power plant for controlling the evaporation rate of a heat recovery boiler that recovers heat from the fuel gas of the coal gasification combined cycle power plant, characterized by measuring the gas temperature at the gas-side outlet of the evaporator of the heat recovery boiler and measuring the circulating water temperature at the circulating water-side inlet of the evaporator of the heat recovery boiler, controlling the evaporator circulation flow rate of the heat recovery boiler so that the actual temperature difference between the gas temperature and the circulating water temperature is close to a temperature difference set value, and pre-controlling the evaporator circulation flow rate of the heat recovery boiler based on a soot blower operation command to remove fuel ash deposited on the surface of the heat transfer tubes of the heat recovery boiler.

[0006] Japanese Registered Patent Publication No. 2011-145827 relates to an information processing device, a method for identifying a connected device, and a program. The information processing device is capable of communicating through a network between a plurality of communication devices. It is characterized by having a storage unit that stores allocation information for connecting a plurality of communication devices and a plurality of processes mapped to the plurality of communication devices to process data transmitted from each of the plurality of communication devices, and a means for acquiring information for specifying a communication device when the plurality of communication devices are connected, and, based on the acquired information, identifying a process corresponding to the communication device from the allocation information stored in the storage unit to create a virtual device file for connecting the connected communication device and the identified process, and a means for establishing a communication link between the process and the connected corresponding communication device based on the created virtual device file.

[0007] Japanese Registered Patent Publication No. 2021-113599 relates to a method for generating teacher data, a method for generating a coke grade prediction model, a coke grade prediction method, a system, and a program. A method for generating teacher data used in machine learning of a prediction model for predicting the grade of coke comprises the steps of: acquiring raw coal physical property data associated with multiple physical property values ​​of each of multiple types of raw coal for each of the multiple types of raw coal; performing dimensionality reduction processing on multiple physical property values ​​of each of the multiple types of raw coal for each of the raw coal physical property data and generating dimensionality-reduced feature data representing each of the multiple physical property values; classifying each of the multiple types of raw coal into one of multiple clusters by cluster analysis of the generated feature data and determining the correspondence relationship between the raw coal and the cluster; acquiring actual measurement data by connecting the blending ratio of each of the multiple types of raw coal and the actually obtained grade of coke; converting the blending ratio of each of the raw coal in the actual measurement data into the blending ratio of each of the clusters using the correspondence relationship between the raw coal and the clusters, and using the converted blending ratio of each of the clusters as input A method for generating teacher data is provided, comprising the step of generating teacher data that outputs the grade value of the above coke.

[0008] Japanese Registered Patent Publication No. 2014-040943 relates to a waste charging control device and a waste charging control method. The waste charging control device for charging waste into a waste melting furnace comprises: a CO2 flow rate measuring means for measuring the flow rate of CO2 contained in the flue gas from the waste melting furnace; a correlation storage means for storing data representing a proportional correlation between the CO2 flow rate and the flue gas flow rate unit; a flue gas flow rate unit calculation means for calculating the flue gas flow rate unit based on the CO2 flow rate based on the data stored in the correlation storage means; an instantaneous waste treatment speed calculation means for calculating the instantaneous waste treatment speed of the waste melting furnace based on the flue gas volume from the waste melting furnace and the flue gas flow rate unit; a waste treatment volume accumulation value calculation means for calculating the accumulation value of the instantaneous waste treatment volume of the waste melting furnace based on the instantaneous waste treatment speed calculated by the instantaneous waste treatment speed calculation means; and a first waste that determines the next waste charging timing by comparing the accumulation value calculated by the waste treatment volume accumulation value calculation means with a predetermined threshold value. A waste charging control device is provided, comprising a means for determining charging timing and a waste charging means for charging the waste into the waste melting furnace according to the determination of the first waste charging timing determination means.

[0009] Typically, in the operation process technology of a coal gasification combined cycle power plant, the fuel determination unit is characterized by determining whether to mix flux (additive) into the coal based on the gasification suitability results and providing an operation guide according to the flux input ratio. It is characterized by selecting a primary target coal from the primary coal types, analyzing plant performance while changing the mixing ratio of secondary target coals at a predetermined rate, and controlling the gasification plant after determining gasification suitability. Since low-quality syngas is produced due to the instability of the initial operation, the above control method requires an operation method suitable for the initial operation.

[0010] Furthermore, conventional technology for coal gasification combined cycle power generation, which uses a mixture of coals with different characteristics, controls the amount of flux, such as fly ash, based on the SiO2, Al2O3, and other components (FeO3, CaO) contained in the coal. Since the primary purpose of this technology is to reduce the cost of input raw materials by adding fly ash and additives, there is a need for technological development that also considers the setting of mixing ratios for coals with different characteristics.

[0011] (Prior Art Literature)

[0012] (Patent Literature)

[0013] Korean Registered Patent Publication No. 10-2200408

[0014] Korean Registered Patent Publication No. 10-2200407

[0015] Japanese Registered Patent Publication No. 6080567

[0016] Japanese Patent Publication No. 2011-145827

[0017] Japanese Patent Publication No. 2021-113599

[0018] Japanese Patent Publication No. 2014-040943

[0019] The object of the present invention is a method for maximizing the operating efficiency of a coal gasification combined cycle power generation system using data labeling automation technology.

[0020] To achieve this purpose, a method for operating a coal gasification combined cycle power plant is provided, comprising: a) securing basic data obtained from one or more raw materials and additives, one or more sensor means installed in the coal gasification combined cycle power plant, and unique values ​​of said raw materials and additives; b) securing process data obtained from said sensor means; c) securing output data obtained through said coal gasification combined cycle power plant; d) combining said basic data, process data, and output data to label the data with a machine learning model; and e) deriving optimal operating conditions based on said labeled data.

[0021] In addition, step d) above may include a step of performing initial machine learning using a pseudo-labeling method and then retraining.

[0022] In addition, the sensor means may be installed at one or more of the inlet / outlet of the individual process forming the coal gasification combined cycle power generation and the system constituting the coal gasification combined cycle power generation.

[0023] In addition, the above basic data may include one or more of the total moisture content, higher heating value, sulfur content, industrial analysis, elemental analysis, ash composition, and ash melting point value.

[0024] In addition, the output data may include one or more of the above-mentioned power, thermal energy, and synthesis gas production amount, and the difference between the target value and the actual value of the output data may be labeled as data.

[0025] In addition, if either of the above 1) cases where multiple types of raw materials are mixed and 2) cases where the difference between the target value and the actual value of the output data exceeds the allowable range of the data is satisfied, the method may additionally include step d-1) of automatically refining and re-labeling the labeled data.

[0026] In addition, step d-2) of generating an alarm based on the above relabeling may be additionally included.

[0027] In addition, the above process data may include one or more of the temperature, pressure, and gas composition obtained from one or more of the gasification, burner, heat exchanger, and turbine equipment.

[0028] In addition, the method may further include a step of storing labeling data and re-labeled data using the above basic data, process data, and output data in a labeling storage means.

[0029] In addition, it further includes a data analysis means for obtaining prediction data through simulation and labeling data analysis to derive the optimal operating conditions, and can provide feedback data analyzed through the data analysis means.

[0030] A coal gasification combined cycle power generation operating device may be provided, comprising: a supply unit for supplying one or more raw materials and additives; one or more sensor means installed in the coal gasification combined cycle power generation system; a basic data collection unit for securing basic data, wherein the basic data is collected based on the unique values ​​possessed by the raw materials and additives; a process data collection unit for securing process data, wherein the process data is collected based on data obtained from the sensor means; an output data collection unit for securing output data, wherein the output data includes results obtained through the coal gasification combined cycle power generation system; a data processing unit that aggregates the basic data, process data, and output data and labels them based on a machine learning model; and an operating condition derivation unit that derives optimal operating conditions based on the labeled data.

[0031] The present invention can also be provided in a form that combines various means for solving the above problem.

[0032] According to the present invention, by labeling data for one or more raw materials and additives, optimal operating conditions can be presented based on the quality of the final production energy source, production information, etc., and plant operation control can also be performed.

[0033] Figure 1 is a schematic diagram illustrating the sequence of a coal gasification combined cycle power generation operation method, which is an embodiment of the present invention.

[0034] Embodiments that enable a person skilled in the art to easily implement the present invention are described in detail below with reference to the attached drawings. However, in describing the operating principles of preferred embodiments of the present invention in detail, if it is determined that a detailed description of related known functions or configurations may unnecessarily obscure the essence of the present invention, such detailed description is omitted.

[0035] In addition, the same reference numerals are used for parts having similar functions and operations throughout the drawings. Throughout the specification, when a part is described as being connected to another part, this includes not only cases where they are directly connected, but also cases where they are indirectly connected with other elements in between. Furthermore, unless specifically stated otherwise, the inclusion of a certain component does not exclude other components but implies that additional components may be included.

[0036] In addition, any limitation or addition to an embodiment in this specification may apply not only to a specific embodiment but also to other embodiments.

[0037] Additionally, throughout the description and claims of the present invention, items indicated in the singular include items indicated in the plural unless otherwise noted.

[0038] The present invention will be described in detail with reference to the drawings and embodiments.

[0039] A person skilled in the art to which this invention pertains will be able to perform various applications and modifications within the scope of this invention based on the above content.

[0040] Figure 1 is a schematic diagram illustrating the sequence of a coal gasification combined cycle power generation operation method, which is an embodiment of the present invention.

[0041] The present invention provides a method for operating a coal gasification combined cycle power plant, comprising: a) securing basic data obtained from one or more raw materials and additives, one or more sensor means installed in the coal gasification combined cycle power plant, and unique values ​​of said raw materials and additives; b) securing process data obtained from said sensor means; c) securing output data obtained through said coal gasification combined cycle power plant; d) labeling data using a machine learning model by combining said basic data, process data, and output data; and e) deriving optimal operating conditions based on said labeled data.

[0042] In addition, step d) above may include a step of performing initial machine learning using a pseudo-labeling method and then retraining.

[0043] In addition, the sensor means may be installed at one or more of the inlet / outlet of the individual process forming the coal gasification combined cycle power generation and the system constituting the coal gasification combined cycle power generation.

[0044] In addition, the above basic data may include one or more of the total moisture content, higher heating value, sulfur content, industrial analysis, elemental analysis, ash composition, and ash melting point value.

[0045] In addition, the output data may include one or more of the above-mentioned power, thermal energy, and synthesis gas production amount, and the difference between the target value and the actual value of the output data may be labeled as data.

[0046] In addition, if either of the above 1) cases where multiple types of raw materials are mixed and 2) cases where the difference between the target value and the actual value of the output data exceeds the allowable range of the data is satisfied, the method may additionally include step d-1) of automatically refining and re-labeling the labeled data.

[0047] In addition, step d-2) of generating an alarm based on the above relabeling may be additionally included.

[0048] In addition, the above process data may include one or more of the temperature, pressure, and gas composition obtained from one or more of the gasification, burner, heat exchanger, and turbine equipment.

[0049] In addition, the method may further include a step of storing labeling data and re-labeled data using the above basic data, process data, and output data in a labeling storage means.

[0050] In addition, it further includes a data analysis means for obtaining prediction data through simulation and labeling data analysis to derive the optimal operating conditions, and can provide feedback data analyzed through the data analysis means.

[0051] A coal gasification combined cycle power generation operating device may be provided, comprising: a supply unit for supplying one or more raw materials and additives; one or more sensor means installed in the coal gasification combined cycle power generation system; a basic data collection unit for securing basic data, wherein the basic data is collected based on the unique values ​​possessed by the raw materials and additives; a process data collection unit for securing process data, wherein the process data is collected based on data obtained from the sensor means; an output data collection unit for securing output data, wherein the output data includes results obtained through the coal gasification combined cycle power generation system; a data processing unit that aggregates the basic data, process data, and output data and labels them based on a machine learning model; and an operating condition derivation unit that derives optimal operating conditions based on the labeled data.

[0052] In addition, the sensor means may include a communication module that collects data in real time and transmits it.

[0053] In addition, the above basic data, process data, and output data can be stored in real time on a cloud server or local storage.

[0054] In addition, the above-mentioned driving condition derivation unit can perform an automatic optimization function that updates the prediction model based on real-time data.

[0055] In addition, the system may include a monitoring module that immediately detects the relevant data and provides a notification to the operator when an abnormal situation occurs.

[0056] In addition, the data processing unit may include data preprocessing functions such as missing value correction, outlier removal, and data normalization.

[0057] In addition, the above machine learning model can perform the function of applying multiple algorithms (deep learning, random forest, support vector machine, etc.) and comparing and verifying their performance.

[0058] In addition, the optimal operating condition derivation unit described above can perform simulations based on various scenarios and predict optimized conditions.

[0059] In addition, the re-labeled data can be stored, and additional insights can be derived through long-term analysis of driving patterns.

[0060] In addition, the system can visually display the operating status, optimal conditions, and analysis results through a user interface (UI).

[0061] In addition, the above system can predict output data based on changes in the composition of raw materials and provide a pre-warning function based on this.

[0062] In addition, the above system may include a function that predicts signs of equipment failure and recommends preventive maintenance through an artificial intelligence (AI) model.

[0063] In addition, the sensor means may include a wireless sensor or an IoT sensor, thereby enabling remote data collection.

[0064] In addition, the above data analysis means can improve power generation efficiency based on past operation data by utilizing big data analysis.

[0065] In addition, the above system can perform an automatic operation function that automatically applies optimal conditions and controls equipment operation without operator intervention.

[0066] In addition, the above output data may enable comprehensive energy management by including linkage data with external systems of the coal gasification combined cycle power generation.

[0067] In addition, the above system can collect and analyze data on greenhouse gas emissions and environmental pollutant emissions to enable compliance with environmental regulations.

[0068] In addition, the data processing unit may include a function to evaluate the reliability of the data and assign weights according to the reliability.

[0069] In addition, the above system can derive optimal operating conditions in real time by considering various external conditions (weather, load changes, etc.).

[0070] In addition, the above data analysis means can analyze the cause when abnormal data occurs and suggest alternative driving strategies.

[0071] In addition, the above system may include a function to automatically generate energy efficiency evaluation results in the form of a report.

[0072] In addition, the data collection unit can integrate various types of sensor data to enable complex data analysis.

[0073] In addition, the optimal driving condition derivation unit described above can utilize digital twin technology to conduct driving simulations in a virtual environment and derive optimized conditions.

[0074] In addition, the above system may include a function to optimize the energy mix in conjunction with renewable energy (solar, wind, etc.).

[0075] In addition, the above output data can quantitatively evaluate greenhouse gas emissions and be linked to a carbon emission trading system.

[0076] In addition, the above system can be linked with an Energy Storage System (ESS) to efficiently store and distribute generated electricity and thermal energy.

[0077] In addition, the above system can be linked with a smart grid system that supplies excess power produced to other industrial facilities or an external power grid.

[0078] In addition, the above data analysis means can implement power usage prediction and demand response (DR) systems to maximize energy efficiency.

[0079] In addition, the above system is linked with a cloud-based big data platform to analyze large-scale data in real time and derive optimal operating conditions.

[0080] Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined by the claims set forth below as well as equivalents thereof.

Claims

1. One or more raw materials and additives; One or more sensor means installed in a coal gasification combined cycle power plant; a) Step of securing basic data obtained from the unique values ​​of the above raw materials and additives; b) Step of securing process data obtained from the sensor means above; c) Step of securing output data obtained through the above coal gasification combined cycle power generation; Step d) of combining the above basic data, process data, and output data to label the data using a machine learning model; and A method for operating a coal gasification combined cycle power plant, comprising: step e) deriving optimal operating conditions based on the above-mentioned labeled data.

2. In Paragraph 1, A method for operating a coal gasification combined cycle power plant, comprising the step of performing initial machine learning using a pseudo-labeling method and then re-learning in step d) above.

3. In Paragraph 1, A method for operating a coal gasification combined cycle power generation in which the sensor means is installed at one or more of the inlet / outlet of an individual process forming the coal gasification combined cycle power generation and the system constituting the coal gasification combined cycle power generation.

4. In Paragraph 1, The above basic data is a method for operating a coal gasification combined cycle power plant that includes one or more of the following: total moisture content, higher heating value, sulfur content, industrial analysis, elemental analysis, ash composition, and ash melting point value.

5. In Paragraph 1, A method for operating a coal gasification combined cycle power generation system that includes at least one of the above output data power, thermal energy, and synthesis gas production, and labels the difference between the target value and the actual value of the above output data as data.

6. In Paragraph 1 A method for operating a coal gasification combined cycle power plant, comprising an additional step d-1) of automatically refining and relabeling labeled data when either of the above conditions is satisfied: 1) when multiple types of raw materials are mixed, or 2) when the difference between the target value and the actual value of the output data exceeds the allowable range of the data.

7. In Paragraph 6 A method for operating a coal gasification combined cycle power plant, additionally including step d-2) of generating an alarm according to the above relabeling.

8. In Paragraph 1 The above process data is a method for operating a coal gasification combined cycle power generation system that includes one or more of the temperature, pressure, and gas composition obtained from one or more of the gasification, burner, heat exchanger, and turbine facilities.

9. In Paragraph 1, A method for operating a coal gasification combined cycle power plant, further comprising the step of storing labeled data and re-labeled data using the above basic data, process data, and output data in a labeling storage means.

10. In Paragraph 1, A method for operating a coal gasification combined cycle power plant, which additionally includes a labeling data analysis for deriving the above optimal operating conditions and a data analysis means for securing prediction data through simulation, and provides feedback data analyzed through the data analysis means.

11. As a device for optimizing the operation of a coal gasification combined cycle power generation system, A supply unit that supplies one or more raw materials and additives; One or more sensor means installed in a coal gasification combined cycle power plant; Basic data collection unit for securing basic data; The above basic data is collected based on the unique values ​​of raw materials and additives, and Process data collection unit for securing process data; The above process data is collected based on data obtained from the sensor means, and Output data collection unit for securing output data; The above output data includes results obtained through coal gasification combined cycle power generation, and A data processing unit that aggregates basic data, process data, and output data and labels them based on a machine learning model; and A coal gasification combined cycle power plant operation device comprising: an operation condition derivation unit that derives optimal operation conditions based on the above-mentioned labeled data.

Images