Boiler control and training method, device and system based on neural network and medium

Abstract

The invention provides a boiler control and training method, device and system based on a neural network and a medium. The control method comprises the steps of obtaining boiler operation data; inputting the boiler operation data into a trained boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion; and obtaining an optimal control scheme of the boiler based on an optimization target of parameters in the prediction result of boiler combustion. According to the method, under the condition that boiler equipment parameters are not changed, the operation efficiency of the boiler is improved, and the fuel consumption is reduced. The problems that on-site boiler debugging is large in repeated trial workload, the debugging effect is poor, and boiler operation is affected are solved. The invention further provides a training method of the boiler control neural network model, and the prediction accuracy of the neural network model is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technologies, and in particular to a boiler control, training method, device, system and medium based on a neural network. Background Art

[0002] With the rapid development of China's industry, the consumption demand for gas has grown rapidly. Gas boilers have great development prospects in the industrial field due to their high combustion efficiency, convenient automatic control, and flexible performance adjustment. Therefore, on the premise of ensuring the safe operation of gas boiler units, it is very important to keep the units in the optimal operation state and minimize fuel consumption. Therefore, it is necessary to study the problem of optimizing the combustion system of gas boilers. Its essence is to adjust the operating parameters of the boiler without changing the equipment parameters of the gas boiler, so that the boiler reaches the best operating conditions, stabilizes the combustion state in the furnace, improves the operating efficiency of the boiler, and reduces the fuel consumption.

[0003] The traditional optimization method is to conduct combustion adjustment tests on site and obtain optimized combustion conditions by means of controlling variables. However, due to the limited operating conditions in on-site commissioning, the combustion conditions of the boiler are related to many factors such as the external environment, boiler load, and fuel type. The traditional optimization scheme may deviate from the best operating conditions or even become invalid under the influence of factors such as changes in the external environment. In the long run, this will not only lead to a decrease in boiler efficiency, but may also cause unstable combustion and accidents.

[0004] Therefore, there is an urgent need for a new optimization method for boiler combustion optimization parameters to improve the combustion efficiency of the boiler while ensuring operation safety. Summary of the Invention

[0005] In view of the above-mentioned disadvantages of the prior art, the purpose of the present disclosure is to provide a boiler control, training method, device, system and medium based on a neural network to solve the problems in the related technologies.

[0006] The first aspect of the present disclosure provides a boiler control method based on a neural network, including: obtaining boiler operation data; inputting the boiler operation data into a trained boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion; and obtaining an optimal control scheme for the boiler based on the optimization objectives of the parameters in the prediction result of the boiler combustion.

[0007] In an embodiment of the first aspect, the boiler operation data includes one or more combinations of the following parameters: boiler load, fuel flow rate, fuel temperature; air supply volume during combustion, opening degree of the guide vane baffle at the outlet of the air supply fan, where the air supply volume includes the primary air supply volume and the secondary air supply volume; and the prediction result of the boiler combustion includes one or more combinations of the following parameters: oxygen content in flue gas, flue gas temperature..

[0008] In an embodiment of the first aspect, it includes: the boiler operation data includes the fuel temperature, and the predicted results of the boiler combustion include: the oxygen content in the flue gas and the flue gas temperature; the optimal ranges of the fuel temperature, the oxygen content in the flue gas, and the flue gas temperature are preset respectively; the boiler operation data is input into a pre-trained boiler control neural network model for analysis and processing to predict the oxygen content in the flue gas and the flue gas temperature after the boiler combustion; when the fuel temperature, the oxygen content in the flue gas, and the flue gas temperature all fall within their respective optimal ranges, the boiler operation data is the optimal control scheme for the boiler.

[0009] In an embodiment of the first aspect, the inputting the boiler operation data into a pre-trained boiler control neural network model for analysis and processing includes: presetting a time range to eliminate the operation data exceeding the time range; presetting an error threshold to eliminate the operation data exceeding the error threshold; presetting multiple value ranges, and retaining at least one operation data within each value range; and normalizing the retained operation data.

[0010] In an embodiment of the first aspect, the formula for normalizing the retained operation data is:

[0011]

[0012] In the formula, is the value of the normalization processing of this kind of data; x is a certain kind of operation data; x max is the maximum value in this kind of operation data; x min is the minimum value in this kind of operation data;

[0013] In an embodiment of the first aspect, the boiler control neural network model includes: an input layer, a hidden layer, and an output layer. Among them, the input layer has five nodes, which respectively receive the boiler flow rate, the primary air supply volume, the secondary air supply volume, the oxygen content in the flue gas, and the flue gas temperature; the output layer has three nodes, which respectively receive the fuel flow rate, the fuel temperature, and the opening degree of the guide vane baffle at the outlet of the air blower.

[0014] In an embodiment of the first aspect, the boiler control neural network model includes: an input layer, a hidden layer, and an output layer. Among them, the input layer has five nodes, which respectively receive the boiler flow rate, the primary air supply volume, the secondary air supply volume, the oxygen content in the flue gas, and the flue gas temperature; the output layer has three nodes, which respectively receive the fuel flow rate, the fuel temperature, and the opening degree of the guide vane baffle at the outlet of the air blower.

[0015] The second aspect of the present disclosure discloses a training method for a boiler control neural network model, including: obtaining the operation data of boiler combustion as the training set of the neural network model; inputting the training set into the boiler control neural network model for analysis and processing to obtain the prediction result of boiler combustion; obtaining the actual operation result of boiler combustion; and determining the training result based on the error between the prediction result and the actual operation result.

[0016] In an embodiment of the second aspect, determining the training result based on the difference between the prediction result and the actual operation result includes: the error calculation formula between the prediction result and the actual operation result is

[0017]

[0018] where Mse is the error; N is the training set; t i is the prediction result of the boiler control neural network model; y i is the actual operation result of the boiler; when Mse is less than the threshold, the boiler control neural network model meets the accuracy requirement and the training is successful.

[0019] The third aspect of the present disclosure discloses a boiler control device based on a neural network, including: a first acquisition module for obtaining boiler operation data; a first analysis module for inputting the boiler operation data into a pre-trained boiler control neural network model for analysis and processing to obtain the prediction result of boiler combustion; and an output module for obtaining the optimal control scheme of the boiler based on the prediction result of boiler combustion.

[0020] The fourth aspect of the present disclosure discloses a training device for a boiler control neural network model, including: a second acquisition module for obtaining the operation data of boiler combustion as the training set of the neural network model; a second analysis module for inputting the training set into the boiler control neural network model for analysis and processing to obtain the prediction result of boiler combustion; a third acquisition module for obtaining the actual operation result of boiler combustion; and a comparison module for determining the training result based on the error between the prediction result and the actual operation result.

[0021] The fifth aspect of the present disclosure discloses an electronic device, which includes: a processor and a memory; wherein, the memory is used to store a computer program; the processor is used to execute the computer program stored in the memory so that the electronic device executes the boiler control method based on a fuzzy neural network described in any embodiment of the first aspect, or the training method for a boiler control neural network model described in any embodiment of the second aspect.

[0022] The fifth aspect of the present disclosure discloses a computer-readable storage medium, on which a computer program is stored. When the program is executed by an electronic device, it implements the boiler control method based on a fuzzy neural network described in any embodiment of the first aspect, or the training method of the boiler control neural network model described in any embodiment of the second aspect.

[0023] The sixth aspect of the present disclosure discloses a boiler control system based on a neural network, including: a data acquisition device for acquiring the operating data of the boiler and the actual operating results of the boiler combustion; a control module communicatively connected to the data acquisition device for receiving various data acquired by the data acquisition device and sending them to the neural network module; and further for controlling the boiler by adjusting the operating parameters of the boiler; the neural network module communicatively connected to the control device for training the boiler control neural network model; and further for analyzing and processing the operating data of the boiler to obtain an optimal control scheme for the boiler; and further for sending the optimal control scheme to the control module so that the control module adjusts the operating parameters of the boiler to the best.

[0024] In the embodiment of the sixth aspect, the data acquisition device includes: a fuel delivery data collector for collecting fuel temperature and fuel flow; a air supply data collector for collecting the primary air supply volume, the secondary air supply volume, and the opening degree of the guide vane baffle at the outlet of the air supply fan; an induced draft data collector for collecting the oxygen content in the flue gas and the flue gas temperature.

[0025] As described above, in the embodiments of the present disclosure, the operating parameters of the boiler are adjusted according to the output result of the boiler control neural network model, so that the boiler reaches the best working condition, realizing the stable combustion state in the furnace, improving the operating efficiency of the boiler, and reducing the fuel consumption without changing the parameters of the boiler equipment. It solves the problems of large amount of repeated commissioning work on-site and poor commissioning effect, which affect the operation of the boiler. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flowchart showing the boiler control method based on a neural network in an embodiment of the present disclosure.

[0027] Figure 2 A schematic diagram showing the boiler combustion device in an embodiment of the present disclosure.

[0028] Figure 3 A flowchart showing the training method of the boiler control neural network model in an embodiment of the present disclosure.

[0029] Figure 4 A schematic diagram showing the modules of the boiler control device based on a neural network in an embodiment of the present disclosure.

[0030] Figure 5Schematic diagram of modules of a training device for a boiler control neural network model in an embodiment of the present disclosure.

[0031] Figure 6 Schematic diagram of the circuit structure of an electronic device in an embodiment of the present disclosure.

[0032] Figure 7 Schematic diagram of the structure of a boiler control system based on a neural network in an embodiment of the present disclosure. Detailed implementation manners

[0033] The following uses specific specific examples to illustrate the implementation manners of the present disclosure. Those skilled in the art can easily understand other advantages and effects of the present disclosure from the information disclosed in the present disclosure. The present disclosure can also be implemented or applied through other different specific implementation manners. Various details in the present disclosure can also be modified or changed according to different viewpoints and application scenarios without departing from the spirit of the present disclosure. It should be noted that, without conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0034] In the description of the present disclosure, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or a group of embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in the present disclosure and the features of different embodiments or examples.

[0035] In addition, the terms "first" and "second" are only used for the purpose of indication, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" can explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "a group" is two or more, unless otherwise specifically defined.

[0036] In order to clearly illustrate the present disclosure, devices irrelevant to the description are omitted, and the same or similar components throughout the specification are given the same reference numerals.

[0037] The technical terms used herein are only for referring to specific embodiments and are not intended to limit the present disclosure. The singular forms used herein also include the plural forms as long as the statements do not clearly indicate the contrary meaning. The meaning of "including" used in the specification is to specify specific features, regions, integers, steps, operations, elements, and / or components, and does not exclude the existence or addition of other features, regions, integers, steps, operations, elements, and / or components.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this disclosure belongs. Terms defined in commonly used dictionaries are further interpreted to have meanings consistent with the relevant technical literature and the currently presented information. As long as they are not defined, they shall not be over-interpreted as ideal or overly formulaic meanings.

[0039] Next, the technical solutions in the embodiments of the present invention will be described in detail with reference to the accompanying drawings in the embodiments of the present invention.

[0040] As Figure 1 shown, a schematic flowchart of a boiler control method based on a neural network in an embodiment of the present disclosure is presented. The method includes steps S11 - S13.

[0041] Step S11: Obtain boiler operation data.

[0042] In some embodiments, the boiler operation data includes one or more combinations of the following parameters: fuel flow rate, fuel temperature; air supply volume during combustion, opening degree of the guide vane baffle at the outlet of the air supply fan, where the air supply volume includes the primary air supply volume and the secondary air supply volume. The predicted results of the boiler combustion include: oxygen content in the flue gas, flue gas temperature.

[0043] Specifically, as Figure 2 shown, the entire boiler combustion device 20 includes: a fuel feeding module 21, an air supply module 22, and an induced draft module 23. The fuel feeding module 21 feeds the fuel required by the boiler into the boiler furnace for combustion; then the air supply module 22 heats the air and divides it into primary air and secondary air to be respectively fed into the furnace for combustion; finally, the induced draft module 23 draws the completely combusted flue gas from the tail of the boiler to the chimney and discharges it into the atmosphere. That is to say, the operation data of the boiler includes: fuel flow rate and fuel temperature during feeding; primary air supply volume and secondary air supply volume, and opening degree of the guide vane baffle at the outlet of the air supply fan during combustion; oxygen content in the flue gas and flue gas temperature after combustion, and all these data are collected by sensors arranged at appropriate positions of the boiler combustion device 20.

[0044] Further, a boiler operation control platform 30 is communicatively connected to the boiler combustion device 20. Through the boiler operation control platform 30, various parameters of the boiler operation can be directly adjusted; the fuel feeding module 21 is controlled to adjust the fuel flow rate and fuel temperature; the air supply module 22 is controlled to adjust the primary air supply volume, primary air supply temperature, secondary air supply volume, and secondary air supply temperature; the induced draft module 23 is controlled to adjust the induced draft volume. At the same time, the boiler operation control platform 30 has a data receiving and transmitting function, receives various data collected by the sensors, and sends these data into the boiler control neural network model; it can also receive the optimized data of the boiler control neural network model.

[0045] Step S12: Input the boiler operation data into the trained boiler control neural network model for analysis and processing to obtain the prediction result of boiler combustion.

[0046] In some embodiments, the analysis and processing of the boiler operation data includes a preliminary screening of the operation data. The screening principles are as follows: the time when the data is obtained should be as recent as possible, that is, to ensure the timeliness of the data; the data should ensure diversity, that is, to ensure that the input and output data of the model are widely distributed within their respective value ranges, and eliminate the data that is significantly deviated from the actual situation; then the data is normalized.

[0047] Specifically, a preset time range is set to eliminate the operation data that exceeds the time range; a preset error threshold is set to eliminate the operation data that exceeds the error threshold; multiple preset value ranges are set, and at least one operation data within each value range is retained; the retained operation data is normalized.

[0048] Further, the normalization formula is:

[0049]

[0050] In the formula, is the value of the normalization processing of this type of data; x is a certain type of operation data; x max is the maximum value of this type of operation data; x min is the minimum value of this type of operation data.

[0051] In some embodiments, the boiler control neural network model includes three layers: an input layer, a hidden layer, and an output layer. Among them, the input layer has five nodes, which respectively receive the boiler flow rate, the primary air supply volume, the secondary air supply volume, the oxygen content in the flue gas, and the flue gas temperature; the output layer has three nodes, which respectively receive the fuel flow rate, the fuel temperature, and the opening degree of the guide vane baffle at the outlet of the air blower.

[0052] That is to say, the boiler control neural network model analyzes and processes the input boiler operation data and the results of boiler combustion, and can output the optimal input parameters for boiler operation under this condition, namely the specific values of fuel flow rate, the fuel temperature, and the opening degree of the outlet guide vane baffle of the forced draft fan.

[0053] In some embodiments, the activation function used by the boiler control neural network model is to provide large-scale non-linearity, enabling the neural network model to approximate any non-linear function.

[0054] Step S13: Obtain the optimal control scheme for the boiler based on the optimization objectives of the parameters in the predicted results of the boiler combustion.

[0055] In some embodiments, the optimization conditions of the boiler control neural network model are: moderate oxygen content in the flue gas, relatively low flue gas temperature, and relatively high temperature of the fuel before entering the furnace.

[0056] Specifically, since the boiler operation data includes the fuel temperature, and the predicted results of the boiler combustion include the oxygen content in the flue gas and the flue gas temperature, the optimal ranges of the fuel temperature, the oxygen content in the flue gas, and the flue gas temperature are preset respectively;

[0057] Input the boiler operation data into the pre-trained boiler control neural network model for analysis and processing to predict the oxygen content in the flue gas and the flue gas temperature after boiler combustion;

[0058] When the fuel temperature, the oxygen content in the flue gas, and the flue gas temperature all fall within their respective optimal ranges, the boiler operation data is the optimal control scheme for the boiler.

[0059] Obtaining the optimal control scheme for the boiler according to the predicted results includes: presetting the optimal ranges of the fuel temperature, the oxygen content in the flue gas, and the flue gas temperature respectively; inputting the boiler operation data into the pre-trained boiler control neural network model for analysis and processing to predict the oxygen content in the flue gas and the flue gas temperature after boiler combustion; when the fuel temperature, the oxygen content in the flue gas, and the flue gas temperature all fall within their respective optimal ranges, the boiler operation data is the optimal control scheme for the boiler.

[0060] Furthermore, input the optimal control scheme obtained by the boiler control neural network model into the boiler operation control platform to adjust the various operation parameters of the boiler, so as to achieve the best operation condition of the boiler without changing the parameters of the boiler equipment.

[0061] To increase the credibility of the optimal control scheme for the boiler output by the boiler control neural network model, it is necessary to examine the accuracy of this neural network model, that is, to confirm whether the predicted results of the boiler control neural network model are accurate. For example Figure 3As shown in the figure, a training method for a boiler control neural network model in an embodiment of the present disclosure is presented, including steps S31 to S34:

[0062] Step S31: Obtain the operating data of boiler combustion as the training set of the neural network model.

[0063] Specifically, monitor the processes of multiple boiler combustions, and record various parameters of the entire combustion process, such as the boiler load, fuel flow rate, fuel temperature; the air supply volume and the opening degree of the outlet guide vane baffle of the air supply fan during combustion; the oxygen content and the flue gas temperature after combustion. To improve the training effect of the boiler control neural network model, the training set can select boilers with various equipment parameters.

[0064] Step S32: Input the training set into the boiler control neural network model for analysis and processing to obtain the prediction results of boiler combustion.

[0065] Step S33: Obtain the actual operating results of boiler combustion.

[0066] Step S34: Determine the training results based on the error between the prediction results and the actual operating results.

[0067] In some embodiments, the generalization mean square error Mse is used as the calculation standard, and the specific calculation formula is

[0068]

[0069] where Mse is the error; N is the training set; t i is the prediction result of the boiler control neural network model; y i is the actual operating result of the boiler;

[0070] When Mse is less than the threshold, the boiler control neural network model meets the accuracy requirements and the training is successful.

[0071] For example, calculating that Mse is less than 0.01 indicates that the boiler control neural network model has a high accuracy and the training is completed.

[0072] Through the above embodiments, the prediction accuracy of the boiler control neural network model is improved, making the optimal control scheme obtained by this model more credible and closer to the actual results.

[0073] As Figure 4 shown, a neural network-based boiler control device 40 in an embodiment of the present disclosure is presented. It should be noted that the principle and technical implementation of the neural network-based boiler control device can refer to the embodiment of the neural network-based boiler control method in the previous embodiment (for example Figure 1 ), so it will not be repeated in this embodiment.

[0074] Specifically, the neural network-based boiler control device 40 includes: a first acquisition module 41, a first analysis module 42, and an output module 43. Among them,

[0075] The first acquisition module 41 is used to obtain boiler operation data;

[0076] The first analysis module 42 is used to input the boiler operation data into a pre-trained boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion;

[0077] The output module 43 is used to obtain an optimal control scheme for the boiler based on the prediction result of boiler combustion.

[0078] As Figure 5 shown, a training device 500 for a boiler control neural network model according to an embodiment of the present disclosure is shown. It should be noted that the principle and technical implementation of the training device for the boiler control neural network model can refer to the embodiment of the training method of the boiler control neural network model in the previous embodiment (for example Figure 3 ), so it will not be repeated in this embodiment.

[0079] Specifically, the training device 50 for the boiler control neural network model includes: a second acquisition module 51, a second analysis module 52, a third acquisition module 53, and a comparison module 54.

[0080] The second acquisition module 51 is used to obtain the operation data of boiler combustion as the training set of the neural network model;

[0081] The second analysis module 52 is used to input the training set into the boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion;

[0082] The third acquisition module 53 is used to obtain the actual operation result of boiler combustion;

[0083] The comparison module 54 is used to determine the training result based on the error between the prediction result and the actual operation result.

[0084] It should be particularly noted that in Figure 4 and Figure 5Each functional module in the embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a program instruction product. The program instruction product includes one or a set of program instructions. When the program instructions are loaded and executed on a computer, the processes or functions according to the present disclosure are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The program instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0085] And, Figure 4 and Figure 5 The devices disclosed in the embodiments can be implemented through other module partitioning methods. The device embodiments shown above are merely illustrative. For example, the partitioning of the modules is only a logical function partitioning. In actual implementation, there can be other partitioning methods. For example, a group of modules or modules can be combined or can be dynamically integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed among each other can be through some interfaces, and the indirect couplings or communication connections of the devices or modules can be in an electrical or other form.

[0086] In addition, Figure 4 and Figure 5 Each functional module and sub-module in the embodiments can be dynamically located in a processing component, or each module can exist physically independently, or two or more modules can be dynamically located in a component. The above-mentioned dynamic component can be implemented in the form of hardware or in the form of a software functional module. When the above-mentioned dynamic component is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk, an optical disc, etc.

[0087] It should be particularly noted that the processes or methods represented by the flowcharts in the above embodiments of the present disclosure can be understood as representing modules, segments, or parts of executable instructions including one or more sets configured to implement specific logical functions or processes. And the scope of the preferred embodiments of the present disclosure includes additional implementations, where the functions can be executed in a substantially simultaneous manner or in the reverse order according to the functions involved, rather than in the order shown or discussed.

[0088] As Figure 6 shown, a schematic structural diagram of an electronic device in an embodiment of the present disclosure is shown.

[0089] The electronic device can execute, by running computer program instructions, as Figure 1 orFigure 3 the method in. Exemplarily, the electronic device may be a distributed computing node system, a server group / server, a desktop computer, a laptop computer, etc., so as to be used to run, for example, Figure 1 the control method in to obtain the optimal control scheme of the boiler based on the result. Or, the electronic device may be a vehicle-mounted controller, or a server / server group in the cloud that communicates remotely with a local terminal, a distributed computing node system, etc. By executing Figure 5 the training method of the boiler control neural network model in, to determine whether the boiler control neural network model is completed based on the difference between the prediction result and the actual operation result of the boiler.

[0090] The electronic device 60 includes a bus 61, a processor 62, and a memory 63. The processor 62 and the memory 63 can communicate through the bus 61. Program instructions may be stored in the memory 63. The processor 62 implements the method steps in the previous embodiments by running the program instructions in the memory 63, such as Figure 1 or Figure 3 the method in.

[0091] The bus 61 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, although only a thick line is used in the figure, it does not mean that there is only one bus or one type of bus.

[0092] In some embodiments, the processor 62 may be implemented as a Central Processing Unit (CPU), a Microcontroller Unit (MCU), a System On Chip, or a Field Programmable Gate Array (FPGA), etc. The memory 63 may include a volatile memory for temporarily storing data when running a program, such as a Random Access Memory (RAM).

[0093] The memory 63 may further include a non-volatile memory for data storage, such as a Read-Only Memory (ROM), a flash memory, a Hard Disk Drive (HDD), or a Solid-State Disk (SSD).

[0094] In some embodiments, the electronic device 60 may further include a communicator 64. The communicator 64 is used for external communication. In a specific example, the communicator 64 may include one or a set of wired and / or wireless communication circuit modules. For example, the communicator 64 may include one or more of, for example, a wired network card, a USB module, a serial interface module, etc. The wireless communication protocols followed by the wireless communication module include, for example, Near Field Communication (NFC) technology, Infrared (IR) technology, Global System for Mobile communications (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Time-Division Code Division Multiple Access (TD-SCDMA), Long Term Evolution (LTE), Bluetooth (BT), Global Navigation Satellite System (GNSS), etc.

[0095] Embodiments of the present disclosure may also provide a computer-readable storage medium, characterized in that program instructions are stored, and the program instructions are run and executed, for example Figure 1 the neural network-based boiler control method in the embodiment, or executed, for example Figure 3 the neural network-based boiler control method in the embodiment.

[0096] That is, the method steps in the above embodiments are implemented as software or computer code that can be stored in a recording medium (such as a CD ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or are implemented as computer code that is originally stored in a remote recording medium or a non-transitory machine-readable medium and downloaded through a network and will be stored in a local recording medium, so that the method represented herein can be stored on such a recording medium for software processing using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA).

[0097] As Figure 7 shown, for the boiler combustion control optimization scenario, embodiments of the present disclosure may also provide a neural network-based boiler control system.

[0098] The neural network-based boiler control system 70 includes:

[0099] A data acquisition device 71 for acquiring the operating data of the boiler and the actual operating results of the boiler combustion.

[0100] In some embodiments, the data acquisition device 71 includes:

[0101] A fuel injection data collector 711 for acquiring fuel temperature and fuel flow rate;

[0102] A primary air supply data collector 712 for acquiring the primary air supply volume, the secondary air supply volume, and the opening degree of the guide vane baffle at the outlet of the air supply fan;

[0103] An induced draft data collector 713 for acquiring the oxygen content in the flue gas and the flue gas temperature.

[0104] A control module 72, communicatively connected to the data acquisition device, for receiving various data acquired by the data acquisition device and sending them to the neural network module; and also for controlling the boiler by adjusting the operating parameters of the boiler.

[0105] The neural network module 73, communicatively connected to the control device, for training the boiler control neural network model; also for analyzing and processing the operating data of the boiler to obtain the optimal control scheme for the boiler; and also for sending the optimal control scheme to the control module so that the control module adjusts the operating parameters of the boiler to the best.

[0106] In summary, the embodiments of the present disclosure provide a neural network-based boiler control, training method, device, system, and medium. By analyzing and processing various data of the boiler combustion through the boiler control neural network model, the operating parameters of the boiler under the optimal working conditions of the boiler are obtained under the equipment parameters, stabilizing the combustion state of the boiler, improving the operating efficiency, and reducing the fuel consumption at the same time. And a training method for the boiler control neural network model is also disclosed, improving the prediction accuracy of the neural network model.

[0107] The above embodiments are only illustrative of the principles and effects of the present disclosure, and are not used to limit the present disclosure. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present disclosure. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed by the present disclosure should still be covered by the claims of the present disclosure.

Claims

1. A boiler control method based on a neural network, characterized in that, include: Obtain boiler operation data; Inputting the boiler operation data into a trained boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion; An optimal control scheme for the boiler is obtained based on the optimization target of the parameters in the prediction result of the boiler combustion.

2. The boiler control method based on neural network according to claim 1, wherein The boiler operation data includes one or more combinations of the following parameters: boiler load, fuel flow, fuel temperature; air volume during combustion, and opening of the blower outlet guide vane damper, wherein the air volume includes primary air volume and secondary air volume; The prediction result of the boiler combustion includes one or more combinations of the following parameters: flue gas oxygen content and flue gas temperature.

3. The boiler control method based on a neural network according to claim 1, wherein include: The boiler operation data includes fuel temperature, and the boiler combustion prediction results include: flue gas oxygen content and flue gas temperature; Preset the optimal ranges of fuel temperature, flue gas oxygen content, and flue gas temperature respectively; Inputting the boiler operation data into a pre-trained boiler control neural network model for analysis and processing to predict the oxygen content and flue gas temperature after boiler combustion; When the fuel temperature, the flue gas oxygen content, and the flue gas temperature all fall within their respective optimal ranges, the boiler operation data is an optimal control solution for the boiler.

4. The boiler control method based on a neural network according to claim 1, wherein The step of inputting the boiler operation data into a pre-trained boiler control neural network model for analysis and processing includes: Preset time range and remove operation data exceeding the time range; Preset an error threshold and eliminate operating data that exceeds the error threshold; Multiple value ranges are preset, and at least one operating data is retained in each value range; Normalize the retained running data.

5. The boiler control method based on a neural network according to claim 4, wherein The formula for normalizing the retained operating data is: Wherein, is the value after normalization of this kind of data; x is a certain kind of operation data; x max is the maximum value in this kind of operation data; x min is the minimum value in this kind of operation data.

6. The boiler control method based on a neural network according to claim 2, wherein The boiler control neural network model includes: an input layer, a hidden layer and an output layer, wherein: The input layer has five nodes, which respectively receive the boiler flow, the primary air supply volume, the secondary air supply volume, the flue gas oxygen content, and the flue gas temperature; The output layer has three nodes, which respectively receive the fuel flow, the fuel temperature, and the opening of the blower outlet guide vane.

7. A training method for a boiler control neural network model, characterized in that include: Obtain boiler combustion operation data as a training set for the neural network model; Inputting the training set into the boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion; Obtain actual operation results of boiler combustion; The training result is determined based on the error between the predicted result and the actual running result.

8. The training method of the boiler control neural network model according to claim 7, characterized in that, Determining the training result based on the difference between the prediction result and the actual operation result includes: The error calculation formula between the predicted result and the actual operation result is: Where Mse is the error; N is the training set; t i is the prediction result of the boiler control neural network model; y i is the actual operation result of the boiler; When Mse is less than the threshold, the boiler control neural network model meets the accuracy requirement and the training is successful.

9. A boiler control device based on a neural network, characterized in that, include: The first acquisition module is used to obtain boiler operation data; A first analysis module is used to input the boiler operation data into a pre-trained boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion; The output module is used to obtain an optimal control plan for the boiler based on the prediction result of the boiler combustion.

10. A training device for a boiler control neural network model, characterized in that, include: The second acquisition module is used to obtain boiler combustion operation data as a training set for the neural network model; A second analysis module, configured to input the training set into the boiler control neural network model for analysis and processing to obtain a prediction result of boiler combustion; A third acquisition module, configured to obtain the actual operation result of boiler combustion; A comparison module, configured to determine a training result based on the error between the prediction result and the actual operation result; 11. An electronic device, characterized in that, The electronic device includes: A processor and a memory; Wherein, the memory is used to store a computer program; The processor is configured to execute the computer program stored in the memory, so that the electronic device executes the boiler control method based on a fuzzy neural network according to any one of claims 1 to 6, or the training method of the boiler control neural network model according to claim 7 or 8.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the electronic device, it implements the boiler control method based on a fuzzy neural network according to any one of claims 1 to 6, or the training method of the boiler control neural network model according to claim 7 or 8.

13. A boiler control system based on a neural network, comprising: A data acquisition device, configured to acquire the operation data of the boiler and the actual operation result of boiler combustion; A control module, communicatively connected to the data acquisition device, configured to receive various data acquired by the data acquisition device and send them to the neural network module; It is further configured to control the boiler by adjusting the operation parameters of the boiler; The neural network module, communicatively connected to the control device, is configured to train the boiler control neural network model; is further configured to analyze and process the operation data of the boiler to obtain an optimal control scheme for the boiler; is further configured to send the optimal control scheme to the control module, so that the control module adjusts the operation parameters of the boiler to the best.

14. The neural network-based boiler control system according to claim 13, wherein The data acquisition device includes: A fuel injection data collector, configured to collect fuel temperature and fuel flow rate; An air supply data collector, configured to collect the primary air supply volume, the secondary air supply volume, and the opening degree of the guide vane baffle at the outlet of the air supply fan; An induced draft data collector, configured to collect the oxygen content in the flue gas and the flue gas temperature.

Images