Coal calorific value monitoring method and system based on unit twinborn model

Abstract

The invention discloses a coal calorific value monitoring method and system based on a unit twinborn model, and the method comprises the steps: building a twinborn mechanism model of an object unit, carrying out the advanced prediction of a function based on the twinborn mechanism model, carrying out the advanced prediction of the exhaust gas temperature of the mechanism model and the exhaust gas temperature of the object unit, and employing a fuzzy control algorithm, and correcting the coal quality lower calorific value Q of the twinning mechanism model in real time, and taking the coal quality lower calorific value Q and the corresponding as-received basis components as coal quality definition information of the twinning mechanism model to form closed-loop correction of the coal quality lower calorific value of the real-time twinning model of the unit, so that the real-time monitoring of the coal quality of the coal-fired unit is realized. The system comprises a first model building module, a second model building module, a first correction module, a data processing module and a real-time monitoring module. According to the invention, real-time monitoring of the coal-fired coal quality of the coal-fired unit is realized.

Description

Technical Field

[0001] The present invention belongs to the field of thermal power digital twin, and in particular relates to a coal calorific value monitoring method and system based on a unit twin model. Background Art

[0002] Affected by comprehensive factors such as coal sources, transportation, and fuel costs, my country's thermal power units generally have a large variety of coal types and qualities. The difference in coal quality seriously affects the powder production performance of the pulverizing system, causing an imbalance in the theoretical amount of air required to burn a unit of pulverized coal, making it extremely difficult to accurately control the key parameters of the unit, and posing a huge challenge to the stable combustion of the boiler at low load or rapidly changing loads, which can easily lead to risks such as ash accumulation and slagging in the furnace and high-temperature corrosion.

[0003] The impacts of coal quality fluctuations on coal-fired units include: increased heat loss, increased equipment wear and tear, increased pollutant emissions, etc. For units with very obvious delays and inertia, it will also weaken the control effect of the boiler-turbine coordinated control system.

[0004] There are many types of online coal quality monitoring devices, such as those based on gamma rays, laser pulses, natural radiation, microwaves and other detection methods, all of which have public application cases. Compared with traditional coal quality chemical analysis methods, online coal quality detection technology can achieve real-time and rapid coal quality analysis, solve the problems of complex processes and delayed results in traditional mining, processing and chemical methods, and provide more timely and objective production guidance. However, there are the following disadvantages: High equipment cost: Advanced online coal quality monitoring devices are usually expensive, including the purchase cost of the equipment itself and subsequent costs such as installation, commissioning, and maintenance. This may be a large economic burden for some companies, limiting its widespread application.

[0005] Technical complexity: These devices involve complex physical and chemical principles and advanced technologies, such as spectral analysis and neutron activation analysis, which require professional technicians to operate, maintain and analyze data. Companies may need to invest additional time and resources to train employees to ensure the normal operation of the equipment and accurate interpretation of the data, otherwise the monitoring effect may be affected due to improper operation or insufficient understanding of the technology.

[0006] Environmental adaptability issues: In industrial environments, there are many unfavorable factors such as dust, vibration, temperature changes, humidity, etc., which may affect the performance and stability of online monitoring devices. For example, dust may adhere to sensors or optical components, affecting measurement accuracy; vibration may cause the mechanical parts of the equipment to loosen or be damaged; changes in temperature and humidity may affect the performance of electronic components. Therefore, corresponding protection and maintenance measures need to be taken to ensure the reliable operation of the equipment in harsh environments, which also increases the cost of using the equipment and the difficulty of maintenance.

[0007] Calibration and calibration requirements: Although online monitoring devices can provide data in real time, regular calibration and calibration are still required to ensure the accuracy and reliability of the data. This requires comparative testing with standard samples or reference methods to adjust the measurement parameters of the equipment. If the calibration and calibration are not timely or accurate, it may lead to increased measurement errors and affect production decisions. In addition, the characteristics of different regions and different types of coal may be different, which also increases the complexity of calibration and calibration.

[0008] Challenges of data accuracy: Despite the continuous development of online monitoring technology, there may still be problems with data accuracy in some cases. For example, for some special coal types, complex coal quality composition or extreme working conditions, the online monitoring device may not be able to fully and accurately reflect the true situation of coal quality. In addition, equipment failures, interference factors, etc. may also cause abnormal or inaccurate data, which requires comprehensive analysis combined with other detection methods and manual judgment. Summary of the invention

[0009] The technical problem to be solved by the present invention is to provide a coal quality calorific value monitoring method and system based on the twin model of the unit. By establishing a twin mechanism model of the object unit, based on the advance prediction function of the mechanism twin model, by advance predicting the exhaust temperature of the mechanism model and the exhaust temperature of the object unit, a fuzzy control algorithm is used to correct the low calorific value Q of the coal quality of the twin mechanism model in real time, and the low calorific value Q of the coal quality and its corresponding received basis components are used as the definition information of the coal quality of the twin mechanism model, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, thereby realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

[0010] In order to achieve the above object, the present invention adopts the following technical solution: A coal quality calorific value monitoring method based on a unit twin model includes: Establish a real-time twin mechanism model of the target unit, and generate an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit; According to the historical data of coal quality testing of the target unit, an AI model of coal low calorific value Q and received basic components is established; Based on the exhaust gas temperature of the advanced prediction mechanism model and the exhaust gas temperature of the target unit, the fuzzy control algorithm is used to realize the real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit; Through the AI ​​model, the real-time coal quality low calorific value Q and its corresponding received basic components are used as the coal quality definition information of the real-time twin mechanism model of the unit; The low calorific value Q of the coal quality and the determined basic components of the coal quality are dynamically input into the real-time twin model of the unit, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, thereby realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

[0011] A further improvement of the present invention is to establish a real-time twin mechanism model of the target unit, and generate an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit, including: The real-time twin mechanism model and the advanced prediction mechanism model of the object unit established are different thread tasks of the same model; The real-time twin mechanism model of the unit synchronously accepts the DCS control instructions of the target unit. The model performs real-time calculations and runs with the target unit. At the same time, the operating conditions are saved every 10 seconds. The stored operating conditions are automatically overwritten and updated in the next cycle. After the operating condition is saved, the operating condition saving completion status is issued. This process is executed in a cycle of 10 seconds; the advanced prediction mechanism model does not accept the DCS control instructions of the target unit. After detecting that the operating condition file of the real-time twin mechanism model of the unit is updated, it is loaded in a cycle at intervals of 10 seconds, and advance calculation is performed for 300 seconds within 10 seconds to realize the advanced prediction function.

[0012] A further improvement of the present invention is that the advance prediction mechanism model does not accept the DCS control command of the target unit. After detecting that the real-time twin mechanism model operating condition file of the unit is updated, it is cyclically loaded at intervals of 10 seconds, and advance calculation is performed for 300 seconds within 10 seconds to realize the advance prediction function, including: The advanced prediction mechanism model does not accept the DCS control instructions of the target unit and waits for the completion of the condition saving. When it detects that the condition saving of the real-time twin mechanism model of the unit is completed, it reads the real-time twin mechanism model condition file and performs 300 seconds of advanced simulation within a 10-second cycle. The system determines whether the 300-second advanced simulation is completed within 10 seconds. If so, the prediction is completed and the next prediction cycle is entered. The prediction results are output to the foreground display. If not, the advanced simulation is interrupted, the advanced simulation time is recorded, and the next prediction cycle is entered. The prediction results are output to the foreground display. The advanced prediction mechanism model is loaded cyclically at intervals of 10 seconds.

[0013] A further improvement of the present invention is to establish an AI model of coal low calorific value Q and received basic components based on the historical data of coal quality testing of the target unit, including: According to the historical data of coal quality testing of the target unit, an AI model of the low calorific value Q of coal and the contents of the received basic components C, H, O, N, S, M, and A is established.

[0014] A further improvement of the present invention is that a fuzzy control algorithm is used based on the exhaust gas temperature of the advanced prediction mechanism model and the exhaust gas temperature of the target unit to achieve real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit, including: Error E uses the exhaust temperature of the advance prediction mechanism model and the exhaust gas temperature of the target unit The difference calculation, E= - , error change using advance prediction mechanism model exhaust temperature Changes and speed of change Classify and establish the real-time twin model coal quality low calorific value Q real-time correction value fuzzy control set to realize the real-time twin mechanism model of the object unit outlet smoke exhaust problem The deviation from the object unit is less than 5°C, thus obtaining the low heating value Q of the coal quality of the real-time twin mechanism model of the object unit, that is, the low heating value of the coal quality of the current unit.

[0015] A further improvement of the present invention is that, through the AI ​​model, the real-time coal quality low calorific value Q and its corresponding received base components are used as coal quality definition information of the real-time twin mechanism model of the unit, including: The definition information of the coal quality of the real-time twin mechanism model of the object unit includes the low calorific value and the received basis components. The initial coal quality definition information of the real-time twin mechanism model adopts the verified coal type. The real-time twin mechanism model is used to obtain the real-time coal quality low calorific value Q. The AI ​​model is used to obtain the real-time coal type low calorific value Q and its corresponding received basis components C, H, O, N, S, M, and A contents of the real-time twin mechanism model of the current object unit. The coal quality definition information of the real-time twin mechanism model of the object unit is updated in real time.

[0016] A coal quality calorific value monitoring system based on a unit twin model comprises: The first model building module builds a real-time twin mechanism model of the target unit, and generates an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit; The second model building module builds an AI model of coal low calorific value Q and received basic components based on the historical data of coal quality testing of the target unit; The first correction module uses a fuzzy control algorithm based on the exhaust temperature of the advanced prediction mechanism model and the exhaust temperature of the target unit to achieve real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit; The data processing module uses the AI ​​model to take the real-time coal quality low calorific value Q and its corresponding received basic components as the coal quality definition information of the real-time twin mechanism model of the unit; The real-time monitoring module dynamically inputs the real-time low calorific value Q of the coal quality and the determined basic components of the coal quality into the real-time twin model of the unit, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, and realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

[0017] A further improvement of the present invention is that in the first model building module, a real-time twin mechanism model of the target unit is established, and an advanced prediction mechanism model based on the current state is generated based on the real-time twin mechanism model of the unit, including: The real-time twin mechanism model and the advanced prediction mechanism model of the object unit established are different thread tasks of the same model; The real-time twin mechanism model of the unit synchronously accepts the DCS control instructions of the target unit. The model performs real-time calculations and runs with the target unit. At the same time, the operating conditions are saved every 10 seconds. The stored operating conditions are automatically overwritten and updated in the next cycle. After the operating condition is saved, the operating condition saving completion status is issued. This process is executed in a cycle of 10 seconds; the advanced prediction mechanism model does not accept the DCS control instructions of the target unit. After detecting that the operating condition file of the real-time twin mechanism model of the unit is updated, it is loaded in a cycle at intervals of 10 seconds, and advance calculation is performed for 300 seconds within 10 seconds to realize the advanced prediction function.

[0018] A further improvement of the present invention is that the advance prediction mechanism model does not accept the DCS control command of the target unit. After detecting that the real-time twin mechanism model operating condition file of the unit is updated, it is cyclically loaded at intervals of 10 seconds, and advance calculation is performed for 300 seconds within 10 seconds to realize the advance prediction function, including: The advanced prediction mechanism model does not accept the DCS control instructions of the target unit and waits for the completion of the condition saving. When it detects that the condition saving of the real-time twin mechanism model of the unit is completed, it reads the real-time twin mechanism model condition file and performs 300 seconds of advanced simulation within a 10-second cycle. The system determines whether the 300-second advanced simulation is completed within 10 seconds. If so, the prediction is completed and the next prediction cycle is entered. The prediction results are output to the foreground display. If not, the advanced simulation is interrupted, the advanced simulation time is recorded, and the next prediction cycle is entered. The prediction results are output to the foreground display. The advanced prediction mechanism model is loaded cyclically at intervals of 10 seconds.

[0019] A further improvement of the present invention is that in the second model building module, an AI model of the low calorific value Q of coal quality and the received basic components is established according to the historical data of coal quality testing of the target unit, including: According to the historical data of coal quality testing of the target unit, an AI model of the low calorific value Q of coal and the contents of the received basic components C, H, O, N, S, M, and A is established.

[0020] Compared with the prior art, the present invention has at least the following beneficial technical effects: Differences in coal quality seriously affect the powder-extracting performance of the pulverizing system, causing an imbalance in the theoretical amount of air required to burn a unit of pulverized coal, bringing great difficulties to the accurate regulation of key parameters of the unit, and posing great challenges to the stable combustion of boilers at low load or rapid load changes, which can easily lead to risks such as ash accumulation and slagging in the furnace, and high-temperature corrosion. The present invention provides a coal quality calorific value monitoring method and system based on the twin model of the unit, which can realize real-time monitoring of the low calorific value of coal quality in coal-fired units. After comparative analysis, the accuracy can reach within 5%, which is an extended application of the twin modeling function of thermal power units, with low implementation and maintenance costs and high accuracy, and provides technical and array support for the digital and intelligent development of thermal power units. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The present invention is a flowchart of a method for monitoring calorific value of coal based on a twin model of a unit.

[0022] Figure 2 A structural schematic diagram of a method for advance prediction of a twin model of a unit according to an implementation mode of the present invention.

[0023] Figure 3 The present invention is a structural block diagram of a coal calorific value monitoring system based on a unit twin model according to an implementation mode of the present invention. DETAILED DESCRIPTION

[0024] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0025] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0026] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0027] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0028] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0029] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0030] Example 1 In the present invention, a twin mechanism model of the object unit is established, based on the advance prediction function of the twin mechanism model, the exhaust temperature of the mechanism model and the exhaust temperature of the object unit are predicted in advance, and a fuzzy control algorithm is used to correct the low calorific value Q of the coal quality of the twin mechanism model in real time, and the low calorific value Q of the coal quality and its corresponding received basis components are used as the definition information of the coal quality of the twin mechanism model, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, thereby realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

[0031] like Figure 1 The flowchart of a method for real-time monitoring of coal calorific value based on advance prediction of a twin model of a unit according to an embodiment of the present invention is schematically shown. The method comprises: Establish a real-time twin mechanism model of the target unit, and generate an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit; According to the historical data of coal quality testing of the target unit, an AI model of coal low calorific value Q and received basic components is established; Based on the exhaust gas temperature of the advanced prediction mechanism model and the exhaust gas temperature of the target unit, the fuzzy control algorithm is used to realize the real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit; Through the AI ​​model, the real-time coal quality low calorific value Q and its corresponding received basic components are used as the coal quality definition information of the real-time twin mechanism model of the unit; The low calorific value Q of the coal quality and the determined basic components of the coal quality are dynamically input into the real-time twin model of the unit, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, thereby realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

[0032] In this embodiment, a real-time twin mechanism model of the target unit is established, and the established model complies with the conservation of mass, conservation of energy, and conservation of momentum.

[0033] Mass Conservation Equation Formula 1 Indicates internal of mass flow.

[0034] Energy Conservation Equation Formula 2 Indicates internal The heat flux heat dissipation source, Represents the thermal conductivity of the fluid.

[0035] Momentum Conservation Equation Formula 3 Represents the dynamic viscosity of the fluid.

[0036] In this embodiment, the control instructions of the twin mechanism model are obtained in real time from the DCS system of the unit. When the unit is at 25%-100% rated load, the parameter accuracy should be guaranteed to be within 5%. The control instructions include at least the following instructions: Feed water flow, coal volume, air volume, total steam turbine valve control instruction, medium pressure valve control instruction, condensate pump frequency, circulating water temperature, circulating water control instruction, high and low water level adjustment instruction, deaerator water level adjustment instruction, each section steam extraction valve control instruction, etc.

[0037] Important parameters used for parameter accuracy evaluation include, but are not limited to, the following: (1) Generator power; (2) Feedwater pump outlet main pipe pressure; (3) Economizer inlet feed water temperature; (4) Economizer inlet feed water pressure; (5) Boiler feed water flow rate; (6) Water wall outlet header temperature; (7) Temperature of the inner wall of the steam-water separator; (8) Lower than the inlet steam temperature; (9) Lower than outlet steam temperature; (10) Steam temperature at the outlet of superheater A; (11) Steam temperature at the outlet of superheater B; (12) Steam temperature at the inlet of superheater secondary desuperheater A; (13) Steam temperature at the inlet of the superheater secondary desuperheater B; (14) Steam temperature at the outlet of superheated secondary desuperheater A; (15) Steam temperature at the outlet of superheated secondary desuperheater B; (16) Higher than the outlet steam temperature; (17)Higher than outlet steam pressure; (18) Main steam pressure (engine side); (19) Main steam temperature; (20) Reheat cold section steam temperature; (21) Reheat cold section steam pressure; (22) Low re-export steam temperature A; (23) Low re-export steam temperature B; (24) High re-inlet steam temperature A; (25) High re-inlet steam temperature B; (26) High re-export steam pressure; (27) High re-export steam temperature; (28) Reheating hot section steam temperature; (29) Reheating hot section steam pressure; (30) Extraction steam pressure of each section; (31) Extraction steam temperature of each section; (32) High and low inlet and outlet temperatures; (33) High and low inlet and outlet pressures; (34) Deaerator pressure; (35) Deaerator temperature; (36) Condensate temperature; (37) Condenser vacuum; (38) Circulating water temperature.

[0038] Example 2 According to the historical data of on-site coal quality testing, the corresponding relationship between the low calorific value Q of coal quality and the received basic components is established, that is, the AI ​​model of the low calorific value Q of coal quality and the received basic components (C, H, O, N, S, ASH, M) under different calorific values The initial definition information of the coal quality of the twin mechanism model of the object unit is verified by the type of coal. The correction and update of the coal quality information is based on the advance prediction function of the twin mechanism model. By predicting the exhaust temperature of the mechanism model and the exhaust temperature of the object unit in advance, the fuzzy control algorithm is used to correct the low calorific value Q of the coal quality of the twin mechanism model in real time. The low calorific value Q of the coal quality and its corresponding received basis components are used as the definition information of the coal quality of the twin mechanism model, forming a closed-loop correction of the low calorific value of the coal quality of the synchronous twin model by the unit data, thereby realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

[0039] Example 3 like Figure 2 As shown in the figure, the real-time twin mechanism model of the object unit and the advanced prediction mechanism model established are different thread tasks of the same model, and the implementation process is as follows: The system instantiates the twin mechanism model of the object unit, generates the real-time twin mechanism model and the advanced prediction mechanism model of the object unit. The real-time twin mechanism model and the advanced prediction mechanism model of the unit are different thread tasks of the same model.

[0040] The real-time twin mechanism model of the unit is connected to the DCS control instructions of the target unit in real time. The model performs real-time operations and runs along with the target unit. The results are output to the foreground display. At the same time, the operating conditions are saved every 10 seconds, and the saved operating conditions are automatically overwritten and updated in the next cycle. After the operating conditions are saved, the status of the saved operating conditions is sent out. This process is executed in a cycle with a period of 10 seconds.

[0041] The lead prediction mechanism model does not accept the DCS control instructions of the target unit and waits for the status of the saved operating conditions. When it monitors the status of the saved operating conditions of the real-time twin mechanism model of the unit, it reads the operating condition file of the real-time twin mechanism model and performs a lead simulation for 300 seconds within a 10-second cycle. The system determines whether the lead simulation for 300 seconds is completed within 10 seconds. If so, the prediction is completed and it enters the next prediction cycle, and the prediction results are output to the foreground display. If not, the lead simulation is interrupted, the lead simulation time is recorded, and it enters the next prediction cycle, and the prediction results are output to the foreground display. The lead prediction mechanism model is loaded in a cycle at intervals of 10 seconds.

[0042] Example 4 Due to the large inertia of boiler combustion and the long time for the change of coal type to be conducted to the tail flue, it is difficult to level. In the embodiment of the present invention, a fuzzy control algorithm is adopted for the exhaust gas temperature based on the lead prediction mechanism model and the exhaust gas temperature of the target unit to realize the correction of the low calorific value Q of the real-time coal type of the real-time twin mechanism model of the target unit.

[0043] According to the exhaust gas temperature of the lead prediction model and the exhaust gas temperature of the unit the difference ΔT = - is used as the error (E), E = ΔT = - ; the change and the change rate of the exhaust gas temperature of the prediction model are classified as the error change (EC) to establish a fuzzy control set for the dynamic correction value of the coal type calorific value Q of the synchronous twin model.

[0044] 1) The classification basis of the error (E) is as follows (E = ΔT = - ): (1) If E <= -20°C, E is negative large (NB); (2) If -20 < E <= -10°C, E is negative medium (NM); (3) If -10 < E <= -5°C, E is negative small (NS); (4) If -5 < E <= 5°C, E is zero (0); (5) If 5 < E <= 10 °C, E is positive small (PS); (6) If 10 < E <= 20 °C, E is positive medium (PM); (7) If E > 20 °C, E is positive big (PB); 2) The error change (EC) is classified using the predicted model of the flue gas temperature change and the change rate as follows: (1) When > 0, the flue gas temperature of the lead prediction model rises; (2) When = 0, the flue gas temperature of the lead prediction model remains unchanged; (3) When < 0, the flue gas temperature of the lead prediction model drops; (4) When > 0, the flue gas temperature of the lead prediction model changes rapidly; (5) When = 0, the flue gas temperature of the lead prediction model changes at a constant speed; (6) When < 0, the flue gas temperature of the lead prediction model changes slowly; 3) The classification of the error change (EC) is based on the following: (1) If "the flue gas temperature of the lead prediction model rises" & "the flue gas temperature of the lead prediction model changes rapidly" = true, EC is negative big (NB); (2) If "the flue gas temperature of the lead prediction model rises" & "the flue gas temperature of the lead prediction model changes at a constant speed" = true, EC is negative medium (NM); (3) If "the flue gas temperature of the lead prediction model rises" & "the flue gas temperature of the lead prediction model changes slowly" = true, EC is negative small (NS); (4) If "the flue gas temperature of the lead prediction model remains unchanged" = true, EC is zero (0); (5) If "the flue gas temperature of the lead prediction model drops" & "the flue gas temperature of the lead prediction model changes slowly" = true, EC is positive small (PS); (6) If "the outlet temperature of the unit heat transfer surface drops" & "the outlet temperature of the unit heat transfer surface changes at a constant speed" = true, EC is positive medium (PM); (7) If "the flue gas temperature of the lead prediction model drops" & "the flue gas temperature of the lead prediction model changes rapidly" = true, EC is positive big (PB).

[0045] 4) According to the above conditions, a fuzzy control rule table of the dynamic correction value of the control signal U, i.e. the calorific value Q of the coal type in the synchronous twin model, can be established.

[0046] Table: Fuzzy control rules for coal calorific value Q correction value of real-time twin model

[0047] Example 5 like Figure 3 As shown, the present invention provides a coal quality calorific value monitoring system based on a unit twin model, comprising: The first model building module builds a real-time twin mechanism model of the target unit, and generates an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit; The second model building module builds an AI model of coal low calorific value Q and received basic components based on the historical data of coal quality testing of the target unit; The first correction module uses a fuzzy control algorithm based on the exhaust temperature of the advanced prediction mechanism model and the exhaust temperature of the target unit to achieve real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit; The data processing module uses the AI ​​model to take the real-time coal quality low calorific value Q and its corresponding received basic components as the coal quality definition information of the real-time twin mechanism model of the unit; The real-time monitoring module dynamically inputs the real-time low calorific value Q of the coal quality and the determined basic components of the coal quality into the real-time twin model of the unit, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, and realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

[0048] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0049] In addition, it should be understood that although this specification is described in accordance with the implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that can be understood by those skilled in the art. The above content is only to illustrate the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution according to the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A coal quality calorific value monitoring method based on a unit twin model, characterized in that: include: Establish a real-time twin mechanism model of the target unit, and generate an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit; According to the historical data of coal quality testing of the target unit, an AI model of coal low calorific value Q and received basic components is established; Based on the exhaust gas temperature of the advanced prediction mechanism model and the exhaust gas temperature of the target unit, the fuzzy control algorithm is used to realize the real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit; Through the AI ​​model, the real-time coal quality low calorific value Q and its corresponding received basic components are used as the coal quality definition information of the unit's real-time twin mechanism model; The low calorific value Q of the coal quality and the determined basic components of the coal quality are dynamically input into the real-time twin model of the unit, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, thereby realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

2. The method for monitoring the calorific value of coal based on the twin model of a unit according to claim 1 is characterized in that: Establish a real-time twin mechanism model of the target unit, and generate an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit, including: The real-time twin mechanism model and the advanced prediction mechanism model of the object unit established are different thread tasks of the same model; The real-time twin mechanism model of the unit synchronously accepts the DCS control instructions of the target unit. The model performs real-time calculations and runs with the target unit. At the same time, the operating conditions are saved every 10 seconds. The stored operating conditions are automatically overwritten and updated in the next cycle. After the operating condition is saved, the operating condition saving completion status is issued. This process is executed in a cycle of 10 seconds; the advanced prediction mechanism model does not accept the DCS control instructions of the target unit. After detecting that the operating condition file of the real-time twin mechanism model of the unit is updated, it is loaded in a cycle at intervals of 10 seconds, and advance calculation is performed for 300 seconds within 10 seconds to realize the advanced prediction function.

3. The method for monitoring the calorific value of coal based on the twin model of a unit according to claim 2 is characterized in that: The advanced prediction mechanism model does not accept the DCS control command of the target unit. After detecting that the real-time twin mechanism model operating condition file of the unit is updated, it is loaded cyclically at intervals of 10 seconds and advanced calculation is performed for 300 seconds within 10 seconds to realize the advanced prediction function, including: The advanced prediction mechanism model does not accept the DCS control instructions of the target unit and waits for the completion of the condition saving. When it detects that the condition saving of the real-time twin mechanism model of the unit is completed, it reads the real-time twin mechanism model condition file and performs 300 seconds of advanced simulation within a 10-second cycle. The system determines whether the 300-second advanced simulation is completed within 10 seconds. If so, the prediction is completed and the next prediction cycle is entered. The prediction results are output to the foreground display. If not, the advanced simulation is interrupted, the advanced simulation time is recorded, and the next prediction cycle is entered. The prediction results are output to the foreground display. The advanced prediction mechanism model is loaded cyclically at intervals of 10 seconds.

4. The method for monitoring the calorific value of coal based on the twin model of a unit according to claim 1 is characterized in that: According to the historical data of coal quality testing of the target unit, an AI model of coal low calorific value Q and received basic components is established, including: According to the historical data of coal quality testing of the target unit, an AI model of the low calorific value Q of coal and the contents of the received basic components C, H, O, N, S, M, and A is established.

5. The method for monitoring the calorific value of coal based on the twin model of a unit according to claim 1 is characterized in that: Based on the exhaust gas temperature of the advanced prediction mechanism model and the exhaust gas temperature of the target unit, the fuzzy control algorithm is used to realize the real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit, including: Error E uses the exhaust temperature of the advance prediction mechanism model and the exhaust gas temperature of the target unit The difference calculation, E= - , error change using advance prediction mechanism model exhaust temperature Changes and speed of change Classify and establish the real-time twin model coal quality low calorific value Q real-time correction value fuzzy control set to achieve the object unit real-time twin mechanism model outlet smoke exhaust problem The deviation from the object unit is less than 5°C, thus obtaining the low heating value Q of the coal quality of the real-time twin mechanism model of the object unit, that is, the low heating value of the coal quality of the current unit.

6. The method for monitoring the calorific value of coal based on the twin model of a unit according to claim 1 is characterized in that: Through the AI ​​model, the real-time coal quality low calorific value Q and its corresponding received base components are used as the coal quality definition information of the real-time twin mechanism model of the unit, including: The definition information of the coal quality of the real-time twin mechanism model of the object unit includes the low calorific value and the received basis components. The initial coal quality definition information of the real-time twin mechanism model adopts the verified coal type. The real-time twin mechanism model is used to obtain the real-time coal quality low calorific value Q. The AI ​​model is used to obtain the real-time coal type low calorific value Q and its corresponding received basis components C, H, O, N, S, M, and A contents of the real-time twin mechanism model of the current object unit. The coal quality definition information of the real-time twin mechanism model of the object unit is updated in real time.

7. A coal quality calorific value monitoring system based on the twin model of the unit, characterized in that: include: The first model building module builds a real-time twin mechanism model of the target unit, and generates an advanced prediction mechanism model based on the current state based on the real-time twin mechanism model of the unit; The second model building module builds an AI model of coal low calorific value Q and received basic components based on the historical data of coal quality testing of the target unit; The first correction module uses a fuzzy control algorithm based on the exhaust temperature of the advanced prediction mechanism model and the exhaust temperature of the target unit to achieve real-time correction of the low calorific value Q of the coal quality of the real-time twin mechanism model of the target unit; The data processing module uses the AI ​​model to take the real-time coal quality low calorific value Q and its corresponding received basic components as the coal quality definition information of the real-time twin mechanism model of the unit; The real-time monitoring module dynamically inputs the real-time low calorific value Q of the coal quality and the determined basic components of the coal quality into the real-time twin model of the unit, forming a closed-loop correction of the low calorific value of the coal quality of the real-time twin model of the unit, and realizing real-time monitoring of the coal quality entering the furnace of the coal-fired unit.

8. The coal quality calorific value monitoring system based on the unit twin model according to claim 7 is characterized in that: In the first model building module, a real-time twin mechanism model of the target unit is established, and an advanced prediction mechanism model based on the current state is generated based on the real-time twin mechanism model of the unit, including: The real-time twin mechanism model and the advanced prediction mechanism model of the object unit established are different thread tasks of the same model; The real-time twin mechanism model of the unit synchronously accepts the DCS control instructions of the target unit. The model performs real-time calculations and runs with the target unit. At the same time, the operating conditions are saved every 10 seconds. The stored operating conditions are automatically overwritten and updated in the next cycle. After the operating condition is saved, the operating condition saving completion status is issued. This process is executed in a cycle of 10 seconds; the advanced prediction mechanism model does not accept the DCS control instructions of the target unit. After detecting that the operating condition file of the real-time twin mechanism model of the unit is updated, it is loaded in a cycle at intervals of 10 seconds, and advance calculation is performed for 300 seconds within 10 seconds to realize the advanced prediction function.

9. The coal quality calorific value monitoring system based on the unit twin model according to claim 8 is characterized in that: The advanced prediction mechanism model does not accept the DCS control command of the target unit. After detecting that the real-time twin mechanism model operating condition file of the unit is updated, it is loaded cyclically at intervals of 10 seconds and advanced calculation is performed for 300 seconds within 10 seconds to realize the advanced prediction function, including: The advanced prediction mechanism model does not accept the DCS control instructions of the target unit and waits for the completion of the condition saving. When it detects that the condition saving of the real-time twin mechanism model of the unit is completed, it reads the real-time twin mechanism model condition file and performs 300 seconds of advanced simulation within a 10-second cycle. The system determines whether the 300-second advanced simulation is completed within 10 seconds. If so, the prediction is completed and the next prediction cycle is entered. The prediction results are output to the foreground display. If not, the advanced simulation is interrupted, the advanced simulation time is recorded, and the next prediction cycle is entered. The prediction results are output to the foreground display. The advanced prediction mechanism model is loaded cyclically at intervals of 10 seconds.

10. The method for monitoring the calorific value of coal based on the twin model of a unit according to claim 7, characterized in that: In the second model building module, based on the historical data of coal quality testing of the target unit, an AI model of coal low calorific value Q and received basic components is established, including: According to the historical data of coal quality testing of the target unit, an AI model of the low calorific value Q of coal and the contents of the received basic components C, H, O, N, S, M, and A is established.

Images