A dynamic regulation and control method for recirculating flue gas of a coal-fired unit

Abstract

The application provides a dynamic regulation and control method for recirculated flue gas of a coal-fired unit, and the method comprises the following steps: obtaining a current flue gas NOx concentration value before a denitration device in a combustion process of the coal-fired unit and a flame combustion image; when the current flue gas NOx concentration value exceeds a preset concentration threshold value and / or the flame combustion deviates from a stable state, obtaining current working state information of the coal-fired unit; inputting the current working state information into a pre-trained BP neural network to obtain an injection position and an injection amount of target recirculated flue gas; and adjusting an output power of an induced draft fan and an opening value of an electric valve of a recirculated flue gas pipeline according to the injection position and the injection amount of the target recirculated flue gas. The method can realize an optimal flue gas circulation arrangement scheme, dynamically regulate and control the injection position and the injection amount of the recirculated flue gas, and improve the flue gas circulation effect, so that the coal-fired unit can be safely and stably operated.

Description

Technical Field

[0001] The present invention relates to the technical field of neural networks, and in particular to a dynamic control method for recirculating flue gas of a coal-fired unit. Background Art

[0002] With the gradual integration of renewable energy into the grid, large coal-fired units will face increasing peak-shaving requirements. This means that coal-fired boilers will frequently need to operate at low load, leading to increased NOx emissions. To prevent excessive NOx emissions from causing serious environmental pollution and harming human health, large coal-fired units currently employ various methods to reduce NOx levels. Flue gas recirculation is a widely used technology. This technology extracts a portion of flue gas from the boiler tail and re-injects it into the boiler. Utilizing the lower temperature and oxygen concentration of the recycled flue gas, it lowers the furnace temperature, suppresses combustion, and thus reduces thermal NOx formation.

[0003] However, in related technologies, while the introduction of flue gas recirculation technology can reduce NOx emissions, it may also bring about problems with unstable flame combustion. For example, when the flue gas recirculation rate increases to a certain level, the flame may pulsate significantly, resulting in combustion surge. In addition, the recirculating flue gas air distribution system generally adopts multi-way air distribution, which can be introduced from above the main combustion zone, or a mixture of primary air, secondary air and burnout air can be introduced, and the input ratio of different branch circulations will have a greater impact on boiler operation. Therefore, according to the real-time operating status of the boiler, formulating the optimal flue gas circulation layout plan and dynamically adjusting the injection position and injection amount of the recirculated flue gas are of great significance for the safe and stable operation of coal-fired units during peak regulation. Summary of the Invention

[0004] The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

[0005] To this end, the first purpose of the present invention is to propose a dynamic control method for recirculating flue gas from a coal-fired unit to reduce the workload of flue gas recirculation and reduce the difficulty of operation, formulate an optimal flue gas circulation layout plan, and dynamically control the injection position and injection amount of recirculated flue gas to improve the flue gas circulation effect and enable the coal-fired unit to operate safely and stably.

[0006] The second object of the present invention is to provide a method and device for dynamically controlling the recirculated flue gas of a coal-fired unit.

[0007] A third object of the present invention is to provide a computer device.

[0008] A fourth object of the present invention is to provide a non-transitory computer-readable storage medium.

[0009] To achieve the above-mentioned objectives, a first embodiment of the present invention provides a method for dynamically controlling recirculated flue gas from a coal-fired unit, comprising:

[0010] Obtain the current flue gas NOx concentration value and flame combustion image in front of the denitrification equipment during the combustion process of the coal-fired unit;

[0011] When the current flue gas NOx concentration value exceeds a preset concentration threshold and / or the flame combustion deviates from a stable state, obtaining current working state information of the coal-fired unit;

[0012] Inputting the current working state information into a pre-trained BP neural network to obtain the injection position and injection amount of the target recirculated flue gas;

[0013] The output power of the induced draft fan and the opening value of the electric valve of the recirculating flue gas pipeline are adjusted according to the target injection position and injection amount of the recirculating flue gas.

[0014] Optionally, in an embodiment of the present invention, the current operating status information of the coal-fired unit includes the coal-fired unit load, excess air coefficient, overburn air ratio, flue gas volume, recirculated flue gas temperature and recirculated flue gas oxygen content.

[0015] Optionally, in an embodiment of the present invention, the target injection position and injection amount of the recirculated flue gas include the recirculated flue gas injection amount of the primary air duct, the secondary air duct, above the main combustion zone, and the burnout air duct.

[0016] Optionally, in an embodiment of the present invention, before inputting the current working state information into a pre-trained BP neural network, the method further includes:

[0017] Acquire a training data set, the training data set including historical NOx concentrations and historical flame combustion images in the flue gas before the denitrification device at different recirculated flue gas injection positions and injection amounts;

[0018] The pre-built initial BP neural network is trained according to the training data set to obtain a trained BP neural network.

[0019] Optionally, in an embodiment of the present invention, before training the pre-built initial BP neural network according to the training data set, the method further includes:

[0020] Preprocessing the flame combustion image to remove fly ash from the flame combustion image;

[0021] Performing channel separation on the pre-processed flame combustion image to convert the single-channel image into a binary image;

[0022] The flame characteristic values ​​of the binary image are calculated, where the characteristic values ​​include a flame center area ratio and an eccentricity of the flame center, where the eccentricity is the distance between the flame center and the furnace center.

[0023] Optionally, in an embodiment of the present invention, training a pre-built initial BP neural network according to the training data set includes:

[0024] Inputting the training data in the training data set into the initial BP neural network, calculating the output parameters corresponding to the training data according to the hidden layer, and calculating the error between the output parameters and the target value;

[0025] If the error is greater than a preset error threshold, the weights from the hidden layer to the output layer are adjusted in sequence, and the values ​​of each node are recalculated;

[0026] Repeat the above process until the error is no greater than the preset error threshold;

[0027] The accuracy of the BP neural network is verified according to the training data set. If the accuracy of the BP neural network exceeds a preset accuracy threshold, the BP neural network training is completed; otherwise, the training process is repeated until the accuracy exceeds the preset accuracy threshold.

[0028] To achieve the above-mentioned purpose, a second embodiment of the present invention provides a dynamic control device for recirculating flue gas from a coal-fired unit, comprising:

[0029] The first acquisition module is used to obtain the current flue gas NOx concentration value and flame combustion image in front of the denitrification equipment during the combustion process of the coal-fired unit;

[0030] A second acquisition module is configured to acquire current operating status information of the coal-fired unit when the current flue gas NOx concentration exceeds a preset concentration threshold and / or the flame combustion deviates from a stable state;

[0031] a processing module, configured to input the current working state information into a pre-trained BP neural network to obtain a target injection position and injection amount of recirculated flue gas;

[0032] The adjustment module is used to adjust the output power of the induced draft fan and the opening value of the electric valve of the recirculating flue gas pipeline according to the injection position and injection amount of the target recirculating flue gas.

[0033] In summary, the present invention provides a method and device for dynamically controlling the recirculated flue gas of a coal-fired unit. The method first obtains the current flue gas NOx concentration value and flame combustion image in front of the denitrification equipment during the combustion process of the coal-fired unit. Then, when the current flue gas NOx concentration value exceeds a preset concentration threshold and / or the flame combustion deviates from the stable state, the current working state information of the coal-fired unit is obtained and input into a pre-trained BP neural network to obtain the target injection position and injection amount of the recirculated flue gas. Finally, the output power of the induced draft fan and the opening value of the electric valve of the recirculated flue gas pipeline are adjusted according to the target injection position and injection amount of the recirculated flue gas. Based on this, the method can use the flue gas NOx concentration value and the flame combustion state as trigger conditions to dynamically control and adjust the injection position and injection amount of the flue gas, so that the coal-fired unit can operate safely and stably. The method also uses the current working state information of the coal-fired unit as input parameters and the target injection position and injection amount of the recirculated flue gas as output parameters to utilize the BP neural network to formulate the optimal flue gas circulation layout plan, reduce the workload of the flue gas recirculation, and improve the flue gas circulation effect.

[0034] To achieve the above-mentioned purpose, the third embodiment of the present invention proposes a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method described in the first embodiment of the present invention is implemented.

[0035] To achieve the above-mentioned purpose, the fourth embodiment of the present invention proposes a non-temporary computer-readable storage medium having a computer program stored thereon, which implements the method described in the first embodiment of the present invention when the computer program is executed by a processor.

[0036] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0038] Figure 1 A schematic flow chart of a method for dynamically controlling recirculated flue gas from a coal-fired unit provided by an embodiment of the present invention;

[0039] Figure 2 A schematic structural diagram of a dynamic control device for recirculating flue gas from a coal-fired unit provided by an embodiment of the present invention;

[0040] Figure 3 A schematic diagram of a recirculating flue gas air distribution system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and are not to be construed as limiting the present invention.

[0042] The following describes a method and apparatus for dynamically controlling recirculated flue gas from a coal-fired unit according to an embodiment of the present invention with reference to the accompanying drawings.

[0043] It should be noted that, in one embodiment of the present invention, a BP neural network can be used to assist in the implementation of this method. Therefore, the BP neural network is introduced. Among them, the back propagation (BP) neural network is a multi-layer feedforward neural network trained according to the error back propagation algorithm. The basic BP algorithm includes two processes: forward propagation of the signal and back propagation of the error. That is, when calculating the error output, it is done from input to output, while the adjustment of weights and thresholds is done from output to input. During forward propagation, the input signal acts on the output node through the hidden layer, and after nonlinear transformation, an output signal is generated. If the actual output does not match the expected output, the error back propagation process is started. Error back propagation is to propagate the output error back to the input layer layer by layer through the hidden layer, and distribute the error to all units in each layer, using the error signal obtained from each layer as the basis for adjusting the weights of each unit. By adjusting the connection strength between input nodes and hidden layer nodes, the connection strength between hidden layer nodes and output nodes, and the threshold, the error is reduced along the gradient direction. After repeated learning and training, the network parameters (weights and thresholds) corresponding to the minimum error are determined, and training is terminated. At this point, the trained neural network can independently process the input information of similar samples and output the information with the minimum error after nonlinear transformation.

[0044] Figure 1 A schematic flow chart of a method for dynamically controlling recirculated flue gas from a coal-fired unit provided in an embodiment of the present invention.

[0045] Step S10, obtaining the current NOx concentration value of the flue gas in front of the denitrification equipment during the combustion process of the coal-fired unit and the flame combustion image.

[0046] Among them, in the embodiment of the present invention, it is possible to X The concentration monitoring device obtains the current flue gas NO before the denitrification equipment X concentration.

[0047] Step S20: When the current flue gas NOx concentration exceeds a preset concentration threshold and / or the flame combustion deviates from a stable state, the current working state information of the coal-fired unit is obtained.

[0048] In the embodiment of the present invention, the current working status information of the coal-fired unit includes the coal-fired unit load, excess air coefficient, burnout air ratio, flue gas volume, recirculated flue gas temperature and recirculated flue gas oxygen content.

[0049] And, in the embodiment of the present invention, it is possible to X The concentration monitoring device determines whether the current flue gas NOx concentration value exceeds a preset concentration threshold. In addition, the flame image monitoring device can determine whether the flame combustion deviates from a stable state.

[0050] Furthermore, in an embodiment of the present invention, the flame combustion image obtained can be preprocessed, and then the flame combustion image can be converted into a binary image using channel separation, and the flame characteristic value can be calculated based on the binary image to determine whether the flame combustion is in a state that deviates from a stable state.

[0051] It should be noted that, in this embodiment, when the current flue gas NOx concentration exceeds a preset concentration threshold and / or the flame combustion deviates from a stable state, the coal-fired unit needs to be regulated, wherein the regulation method needs to be implemented by training a BP neural network.

[0052] And, the training of the BP neural network includes:

[0053] Obtain a training data set, which includes historical NOx concentrations and historical flame combustion images in the flue gas before the denitrification equipment at different recirculated flue gas injection positions and injection amounts;

[0054] The pre-built initial BP neural network is trained according to the training data set to obtain a trained BP neural network.

[0055] Furthermore, the method for generating the training data set may further include:

[0056] Preprocessing the flame combustion image to remove fly ash from the flame combustion image;

[0057] Perform channel separation on the preprocessed flame combustion image and convert the single-channel image into a binary image;

[0058] Calculate the flame eigenvalues ​​of the binary image. The eigenvalues ​​include the flame center area ratio and the eccentricity of the flame center. The eccentricity is the distance between the flame center and the furnace center.

[0059] Afterwards, a database is established by combining the flame characteristic values ​​with the current working status information of the coal-fired unit, and the data in the database are initially screened and normalized to obtain a training data set for the BP neural network.

[0060] Furthermore, training the pre-built initial BP neural network according to the training data set may include:

[0061] Input the training data in the training data set into the initial BP neural network, calculate the output parameters corresponding to the training data according to the hidden layer, and calculate the error between the output parameters and the target value;

[0062] If the error is greater than the preset error threshold, the weights from the hidden layer to the output layer are adjusted in sequence, and the values ​​of each node are recalculated;

[0063] Repeat the above process until the error is no greater than the preset error threshold;

[0064] The accuracy of the BP neural network is verified according to the training data set. If the accuracy of the BP neural network exceeds the preset accuracy threshold, the BP neural network training is completed; otherwise, the training process is repeated until the accuracy exceeds the preset accuracy threshold.

[0065] Step S30: input the current working state information into the pre-trained BP neural network to obtain the target injection position and injection amount of the recirculated flue gas.

[0066] Among them, in the embodiment of the present invention, the target injection position and injection amount of the recirculated flue gas include the recirculated flue gas injection amount of the primary air duct, the secondary air duct, above the main combustion zone, and the burnout air duct.

[0067] Specifically, in the embodiment of the present invention, the current working state information can be used as input parameters, and the coal-fired unit load, excess air coefficient, burnout air ratio, flue gas volume, recirculated flue gas temperature and recirculated flue gas oxygen content are set as x1 to x6 respectively, the injection position and injection amount of the target recirculated flue gas are used as output parameters, and the recirculated flue gas injection amount of the primary air duct, secondary air duct, above the main combustion zone, and burnout air duct are set as y1 to y4 respectively. The specific values ​​of the six input parameters are input into the BP neural network respectively to obtain the NO in the flue gas before the denitrification equipment. X The injection position and injection amount of the target recirculated flue gas corresponding to the lowest concentration.

[0068] Step S40 , adjusting the output power of the induced draft fan and the opening value of the electric valve of the recirculating flue gas pipeline according to the target injection position and injection amount of the recirculating flue gas.

[0069] Among them, in the embodiment of the present invention, according to the obtained NO in the flue gas before the denitrification equipment X The injection position and injection amount of the target recirculated flue gas corresponding to the lowest concentration can be set by setting the induced draft fan power, adjusting the total amount of the recirculated flue gas pipeline, and then adjusting the initial value of the electric valve opening of the recirculated flue gas pipeline at the four branches. XThe concentration is dynamically adjusted to adjust the opening of the recirculation flue gas pipe to achieve the lowest NO concentration under stable combustion conditions. X emission.

[0070] In summary, the present invention provides a method for dynamically controlling the recirculated flue gas of a coal-fired unit. The method first obtains the current flue gas NOx concentration value and flame combustion image in front of the denitrification equipment during the combustion process of the coal-fired unit. Then, when the current flue gas NOx concentration value exceeds a preset concentration threshold and / or the flame combustion deviates from the stable state, the current working state information of the coal-fired unit is obtained and input into a pre-trained BP neural network to obtain the target injection position and injection amount of the recirculated flue gas. Finally, the output power of the induced draft fan and the opening value of the electric valve of the recirculated flue gas pipeline are adjusted according to the target injection position and injection amount of the recirculated flue gas. Based on this, the method can use the flue gas NOx concentration value and the flame combustion state as trigger conditions to dynamically control and adjust the injection position and injection amount of the flue gas, so that the coal-fired unit can operate safely and stably. The method also uses the current working state information of the coal-fired unit as input parameters and the target injection position and injection amount of the recirculated flue gas as output parameters to utilize the BP neural network to formulate the optimal flue gas circulation layout plan, reduce the workload of the flue gas recirculation, and improve the flue gas circulation effect.

[0071] Figure 2 This is a schematic structural diagram of a dynamic control device for recirculating flue gas from a coal-fired unit provided by an embodiment of the present invention.

[0072] like Figure 2 As shown, the device includes the following modules:

[0073] The first acquisition module 100 is used to obtain the current NOx concentration value of the flue gas in front of the denitrification equipment and the flame combustion image during the combustion process of the coal-fired unit;

[0074] The second acquisition module 200 is used to obtain the current working state information of the coal-fired unit when the current flue gas NOx concentration value exceeds a preset concentration threshold and / or the flame combustion deviates from the stable state;

[0075] The processing module 300 is used to input the current working state information into the pre-trained BP neural network to obtain the injection position and injection amount of the target recirculated flue gas;

[0076] The adjustment module 400 is used to adjust the output power of the induced draft fan and the opening value of the electric valve of the recirculation flue gas pipeline according to the target injection position and injection amount of the recirculation flue gas.

[0077] It should be noted that the above explanation of the embodiment of the method for regulating the injection position and injection amount of recycled flue gas based on the BP neural network is also applicable to the device of this embodiment. Please refer to the relevant description of the above embodiment and will not be repeated here.

[0078] In summary, the present invention provides a dynamic control device for recirculating flue gas from a coal-fired unit. The device first obtains the current flue gas NOx concentration value and flame combustion image in front of the denitrification equipment during the combustion process of the coal-fired unit. Then, when the current flue gas NOx concentration value exceeds a preset concentration threshold and / or the flame combustion deviates from the stable state, the current working state information of the coal-fired unit is obtained and input into a pre-trained BP neural network to obtain the target injection position and injection amount of the recirculated flue gas. Finally, the output power of the induced draft fan and the opening value of the electric valve of the recirculating flue gas pipeline are adjusted according to the target injection position and injection amount of the recirculated flue gas. Based on this, the method can use the flue gas NOx concentration value and the flame combustion state as trigger conditions to dynamically control and adjust the injection position and injection amount of the flue gas, so that the coal-fired unit can operate safely and stably. The method also uses the current working state information of the coal-fired unit as an input parameter and the target injection position and injection amount of the recirculated flue gas as an output parameter to utilize the BP neural network to formulate an optimal flue gas circulation layout plan, reduce the workload of the flue gas recirculation, and improve the flue gas circulation effect.

[0079] It should be noted that, in the embodiment of the present invention, the dynamic control method of the recirculated flue gas of the coal-fired unit can have a recirculated flue gas air distribution system corresponding to the method. In order to more clearly explain the recirculated flue gas air distribution system in the present invention, the following is a Figure 3 As an illustration.

[0080] Figure 3 This is a schematic diagram of a recirculating flue gas distribution system provided by an embodiment of the present invention. Figure 3 As shown, the system includes: 1, boiler 2, chimney 3, induced draft fan 4, recirculating flue gas pipeline 5, electric valve 6, primary air pipeline 7, secondary air pipeline 8, burnout air pipeline 9, pulverized coal burner 10, flame image monitoring device 11, air duct nozzle 12, NO X Concentration monitoring device.

[0081] In the embodiment of the present invention, when the boiler 1 is in the actual combustion process, the flue gas will pass through NO X The concentration monitoring device 12 moves toward the chimney 2. During this process, if NO X The concentration monitoring device 12 detects NO in the flue gas before the denitrification equipment. XIf the concentration exceeds the standard, or if the flame image monitoring device 10 detects that the flame combustion image deviates from the stable state, the overburn air dynamic control system is triggered. This system automatically inputs the actual parameters under the current operating conditions and, using the established BP neural network model, calculates the optimal recirculated flue gas injection position and injection rate. It then automatically adjusts the power of the induced draft fan 3, the total amount of recirculated flue gas in the recirculated flue gas duct 4, and the electric valves 5 on the four branch recirculating flue gas ducts. Finally, it adjusts the recirculated flue gas injection rate in each branch. The recirculated flue gas control system has four branches, each with an electric valve 5. Furthermore, in the two branches near the bottom of the system, the recirculated flue gas will be mixed with the primary air duct 6 and the secondary air duct 7, respectively, and ejected through the pulverized coal burner 9 to participate in the pulverized coal combustion process. In the remaining two branches, the recirculated flue gas will enter the overburned air duct 8 and mix with the overburned air before being ejected, or directly injected into the furnace through the air duct nozzle 11.

[0082] Furthermore, during the adjustment process, the flame image monitoring device 10 continuously monitors in real time and determines whether the flame image characteristic values ​​are within a specified range. This ensures that while the boiler flame remains stably burning, the NOx concentration detected in real time by the NOx concentration monitoring device 12 is reduced to within the permitted range, thereby ensuring dynamic regulation of NOx levels during the combustion process of the coal-fired unit.

[0083] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0084] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0085] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention pertain.

[0086] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0087] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0088] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0089] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium.

[0090] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and are not to be construed as limiting the present invention. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for dynamically controlling recirculated flue gas from a coal-fired unit, characterized in that: include: Obtain the current flue gas NOx concentration value and flame combustion image in front of the denitrification equipment during the combustion process of the coal-fired unit; When the current flue gas NOx concentration value exceeds a preset concentration threshold and / or the flame combustion deviates from a stable state, obtaining current working state information of the coal-fired unit; Inputting the current working state information into a pre-trained BP neural network to obtain the injection position and injection amount of the target recirculated flue gas; The output power of the induced draft fan and the opening value of the electric valve of the recirculating flue gas pipeline are adjusted according to the target injection position and injection amount of the recirculating flue gas.

2. The dynamic control method according to claim 1, characterized in that: The current working status information of the coal-fired unit includes the coal-fired unit load, excess air coefficient, overburn air ratio, flue gas volume, recirculated flue gas temperature and recirculated flue gas oxygen content.

3. The dynamic control method according to claim 1, wherein: The target injection position and injection amount of the recirculated flue gas include the recirculated flue gas injection amount of the primary air duct, the secondary air duct, above the main combustion zone, and the burnout air duct.

4. The dynamic control method according to any one of claims 1 to 3, characterized in that: Before inputting the current working state information into the pre-trained BP neural network, the method further includes: Acquire a training data set, the training data set including historical NOx concentrations and historical flame combustion images in the flue gas before the denitrification device at different recirculated flue gas injection positions and injection amounts; The pre-built initial BP neural network is trained according to the training data set to obtain a trained BP neural network.

5. The dynamic control method according to claim 4, characterized in that: Before training the pre-built initial BP neural network according to the training data set, the method further includes: Preprocessing the flame combustion image to remove fly ash from the flame combustion image; Performing channel separation on the pre-processed flame combustion image to convert the single-channel image into a binary image; The flame characteristic values ​​of the binary image are calculated, where the characteristic values ​​include a flame center area ratio and an eccentricity of the flame center, where the eccentricity is the distance between the flame center and the furnace center.

6. The dynamic control method according to claim 4, characterized in that: The pre-built initial BP neural network is trained according to the training data set, including: Inputting the training data in the training data set into the initial BP neural network, calculating the output parameters corresponding to the training data according to the hidden layer, and calculating the error between the output parameters and the target value; If the error is greater than a preset error threshold, the weights from the hidden layer to the output layer are adjusted in sequence, and the values ​​of each node are recalculated; Repeat the above process until the error is no greater than the preset error threshold; The accuracy of the BP neural network is verified according to the training data set. If the accuracy of the BP neural network exceeds a preset accuracy threshold, the BP neural network training is completed; otherwise, the training process is repeated until the accuracy exceeds the preset accuracy threshold.

7. A dynamic control device for recirculating flue gas from a coal-fired unit, characterized in that: The first acquisition module is used to obtain the current flue gas NOx concentration value and flame combustion image in front of the denitrification equipment during the combustion process of the coal-fired unit; A second acquisition module is configured to acquire current operating status information of the coal-fired unit when the current flue gas NOx concentration exceeds a preset concentration threshold and / or the flame combustion deviates from a stable state; a processing module, configured to input the current working state information into a pre-trained BP neural network to obtain a target injection position and injection amount of recirculated flue gas; The adjustment module is used to adjust the output power of the induced draft fan and the opening value of the electric valve of the recirculating flue gas pipeline according to the injection position and injection amount of the target recirculating flue gas.

8. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images