A closed loop control method and apparatus for an electrostatic precipitator

Abstract

The application discloses a closed-loop control method and device of an electric dust collector. The closed-loop control method of the electric dust collector comprises the following steps: obtaining boiler data from a distributed control system (DCS) control cabinet through a communication card; obtaining dust removal data from a data acquisition embedded system (DAU) device, wherein the dust removal data comprises electric field parameters and rapping parameters of multiple electric fields; integrating the boiler data and the dust removal data according to a preset frequency, and generating access data; constructing a rapping parameter prediction model of the multiple electric fields according to the access data; and controlling the electric dust collector to perform rapping dust removal by using the rapping parameters output by the rapping parameter prediction model. Therefore, the operation intensity of an operator can be effectively reduced, and energy saving and environmental protection are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of environmental protection and energy saving, and particularly relates to a closed-loop control method and device for an electric dust collector. BACKGROUND

[0002] The rapping control of the electric dust collector is mostly based on manual operation, which not only increases the operation difficulty of the operator but also causes operation failure due to the uneven experience of the operator. Therefore, how to reduce the operation intensity of the operator and save energy is a problem to be solved. SUMMARY

[0003] Therefore, the main purpose of the present application is to provide a closed-loop control method and device for an electric dust collector, which can effectively reduce the operation intensity of the operator and save energy.

[0004] The first aspect of the present application provides a closed-loop control method for an electric dust collector, which comprises:

[0005] obtaining boiler data from a distributed control system (DCS) control cabinet through a communication card;

[0006] obtaining dust removal data from a data acquisition embedded system (DAU) device, wherein the dust removal data comprises electric field parameters and rapping parameters of a plurality of electric fields;

[0007] integrating the boiler data and the dust removal data according to a preset frequency, and generating access data;

[0008] constructing a rapping parameter prediction model of the plurality of electric fields according to the access data;

[0009] controlling the electric dust collector to perform rapping dust removal using the rapping parameters output by the rapping parameter prediction model.

[0010] In some implementations of the first aspect of the present application, the construction of the rapping parameter prediction model of the plurality of electric fields according to the access data can specifically comprise:

[0011] calculating the total dust amount of the boiler flue gas according to the boiler data;

[0012] calculating the dust removal efficiency according to the dust removal data;

[0013] calculating the theoretical dust amount of each electric field in the plurality of electric fields according to the total dust amount of the boiler flue gas and the dust removal efficiency;

[0014] generating working condition data according to the total dust amount of the boiler flue gas, the dust removal efficiency and the theoretical dust value;

[0015] obtaining a neural network model of the rapping parameters of the plurality of electric fields according to the access data and the working condition data.

[0016] In some implementations of the first aspect of the application, the method can further include:

[0017] Filtering the working condition data according to the correlation coefficient of the real values of the beating parameters of the multiple electric fields to obtain working condition parameters, the working condition parameters including: inlet dust concentration, boiler load, flue gas pressure, flue gas temperature, energy consumption of the multiple electric fields, theoretical dust amount, and dust concentration at the dust removal outlet;

[0018] Selecting a network depth parameter and a number of neurons of a neural network model according to the working condition parameters, and generating a beating parameter predicted value;

[0019] Determining a loss function according to the real value of the beating parameter and the beating parameter predicted value.

[0020] Training the beating parameter prediction model according to the access data until the loss function value of the loss function converges, and stopping updating the working condition parameters.

[0021] In some implementations of the first aspect of the application, the method can further include:

[0022] Building a beating parameter compensation model of the multiple electric fields;

[0023] Inputting the beating parameter predicted value into the beating parameter compensation model for compensation, and obtaining a dust removal dust real value;

[0024] Inputting the dust removal dust real value and a preset dust setting value into the beating parameter compensation model to obtain an error compensation value;

[0025] Compensating the beating parameter predicted value according to the error compensation value to obtain a compensated beating parameter predicted value.

[0026] In some implementations of the first aspect of the application, the method can further include:

[0027] When the access closed-loop control appears an abnormal condition, an alarm information is sent to the electric dust collector, and the alarm information is displayed.

[0028] In some implementations of the first aspect of the application, when the access closed-loop control appears an abnormal condition, an alarm information is sent to the electric dust collector, and the alarm information is displayed.

[0029] When the dust emission value at the dust removal outlet of the electric dust collector is greater than a first threshold value, a first-level alarm is performed;

[0030] When the dust emission value at the dust removal outlet of the electric dust collector is greater than a second threshold value, a second-level alarm is performed;

[0031] When the dust emission value at the outlet of the chimney is greater than a third threshold value, an alarm information is sent to the electric dust collector;

[0032] The priority of the first-level alarm is higher than the priority of the second-level alarm.

[0033] In some implementations of the first aspect of the application, the method can further include:

[0034] switching the current operation mode of the electric dust collector to the manual operation mode according to the alarm information.

[0035] The second aspect of the application provides a closed-loop control device of an electric dust collector, which comprises a data interaction system, an intelligent control server, a communication card, and an electric dust collector.

[0036] The data interaction system is configured to obtain boiler data from a distributed control system (DCS) control cabinet through the communication card.

[0037] The data interaction system is further configured to obtain dust removal data from a data acquisition embedded system (DAU) device, wherein the dust removal data comprises electric field parameters and rapping parameters of multiple electric fields.

[0038] The intelligent control server is configured to integrate the boiler data and the dust removal data according to a preset frequency, and generate access data.

[0039] The intelligent control server is further configured to construct a rapping parameter prediction model of the multiple electric fields according to the access data.

[0040] The intelligent control server is further configured to control the electric dust collector to perform rapping dust removal using the rapping parameters output by the rapping parameter prediction model.

[0041] The third aspect of the application provides another closed-loop control device of an electric dust collector, which comprises:

[0042] The acquisition module is configured to obtain boiler data from a distributed control system (DCS) control cabinet through the communication card.

[0043] The acquisition module is configured to obtain dust removal data from a data acquisition embedded system (DAU) device, wherein the dust removal data comprises electric field parameters and rapping parameters of multiple electric fields.

[0044] The generation module is configured to integrate the boiler data and the dust removal data according to a preset frequency, and generate access data.

[0045] The model construction module is configured to construct a rapping parameter prediction model of the multiple electric fields according to the access data.

[0046] The rapping module is configured to control the electric dust collector to perform rapping dust removal using the rapping parameters output by the rapping parameter prediction model.

[0047] The fourth aspect of the present application provides a computer device, characterized in that the device comprises a memory and a processor, and the processor is configured to execute a program stored in the memory to implement the method according to any one of the preceding first aspect.

[0048] Compared with the prior art, the technical scheme provided by the present application has the following beneficial effects:

[0049] The present application obtains boiler data from a distributed control system (DCS) control cabinet through a communication card; obtains dust removal data from a data acquisition embedded system (DAU) device, the dust removal data including electric field parameters and rapping parameters of multiple electric fields; integrates the boiler data and the dust removal data according to a preset frequency, and generates access data; constructs a rapping parameter prediction model of the multiple electric fields according to the access data, and uses the rapping parameter output by the rapping parameter prediction model to control the electric dust collector to perform rapping dust removal. Compared with the prior art scheme which adopts a manual control mode, the operation difficulty of the operating personnel is reduced. Moreover, the rapping parameter prediction model generated according to the working condition parameters and other data can automatically change strategies under different working condition environments, so as to realize dynamic adjustment, which is more convenient for the operating personnel to operate. BRIEF DESCRIPTION OF DRAWINGS

[0050] Figure 1 A flowchart of an electric dust collector closed-loop control method provided by an embodiment of the present application is shown in the figure;

[0051] Figure 2 A flowchart of another electric dust collector closed-loop control method provided by an embodiment of the present application is shown in the figure;

[0052] Figure 3 A flowchart of another electric dust collector closed-loop control method provided by an embodiment of the present application is shown in the figure;

[0053] Figure 4 A flowchart of another electric dust collector closed-loop control method provided by an embodiment of the present application is shown in the figure;

[0054] Figure 5 A training flowchart of a rapping parameter prediction model provided by an embodiment of the present application is shown in the figure;

[0055] Figure 6 A training flowchart of a rapping parameter compensation model provided by an embodiment of the present application is shown in the figure;

[0056] Figure 7 A structural diagram of an electric dust collector closed-loop control device provided by an embodiment of the present application is shown in the figure;

[0057] Figure 8 A structural diagram of another electric dust collector closed-loop control device provided by an embodiment of the present application is shown in the figure;

[0058] Figure 9 A structural schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work are within the scope of protection of the present application.

[0060] The terms "first", "second", "third", "fourth" and the like in the specification and claims of the present application and the above-described drawings, if any, are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to the process, method, product or device.

[0061] Related term explanation:

[0062] Data interaction system: used for automatic switching through logical implementation mode. For example, switching between manual mode / automatic mode through preset logic. In manual mode, the rapping strategy can be set manually, and in automatic mode, the instructions sent by the intelligent control server can be received for optimized automatic control;

[0063] Intelligent control server: used for controlling the operation of the rapping system of the electric precipitator, for example, a neural network model and an optimization compensation model can be used to realize precise rapping of the electric precipitator and realize the algorithm of closed-loop control;

[0064] DAU data acquisition: the acquisition server is connected to the DAU switch or DAU directly through a network cable, and the key parameters of dust removal are obtained and controlled through Modbus TCP;

[0065] IO communication card: also known as communication card, used for data transfer of boiler side and flue gas system related indicators between DCS system and acquisition server, playing a role of connecting the upper and lower parts, facilitating I / O expansion;

[0066] DCS system (Distributed Control System): distributed control system.

[0067] In the prior art, the operating personnel sets the rapping period, rapping time, rapping height and the like based on experience, and after the setting is completed, the parameters remain unchanged for a long time, or the operating personnel adjusts the corresponding parameters in response to the corresponding phenomena after the occurrence of phenomena such as secondary current, secondary voltage characteristic change and dust removal efficiency reduction. However, manual operation is time-consuming and laborious, and due to uneven experience, there may be operation errors. On the other hand, manual control cannot achieve real-time control and may have a large delay. The present application performs rapping control based on a rapping parameter prediction model, and can realize automatic closed-loop control of the electric dust collector.

[0068] Please refer to Figure 1 Since the prior art cannot achieve closed-loop automatic control by using manual methods, the embodiment of the present application provides a closed-loop control method of an electric dust collector, including the following steps:

[0069] S101: Obtain boiler data from a distributed control system (DCS) control cabinet through a communication card;

[0070] It should be noted that the boiler data can include boiler load, original flue gas flow, original flue gas temperature, original flue gas pressure, net flue gas flow, net flue gas temperature and net flue gas pressure and the like related to the boiler side. The communication card can be an IO communication card, which transmits data of boiler and flue gas system related indicators between a distributed control system (DCS) system and a data acquisition server, plays a role of connecting the upper and lower parts, and facilitates I / O expansion.

[0071] S102: Obtain dust removal data from a data acquisition embedded system (DAU) device, the dust removal data including electric field parameters and rapping parameters of a plurality of electric fields;

[0072] It should be noted that the dust removal data can be obtained from the data acquisition embedded system DAU device, and the dust removal data can also be obtained from the switch. The electric field parameters can include secondary current, secondary voltage, spark rate and other data related to the electric field of each electric field; the rapping parameters can include rapping period, rapping time, rapping height and rapping frequency and other data related to the rapping device. Among them, the spark rate refers to the discharge frequency of the electric field, and the general adjustment method is to reduce the secondary voltage and the secondary current. The principle of electrostatic dust removal is that the high voltage between the grounded anode plate and the cathode frame forms a corona discharge, ionizes the air to generate charged ions attached to the dust, and the charged ions are forced to approach the anode plate in the electric field between the anode plate and the cathode frame and are concentrated on the anode plate. The dust is shaken off by the rapping device and collected in the ash bucket. The spark is generated by the discharge between the anode plate and the cathode frame, so sufficient insulation and discharge conditions are required between the anode and the cathode. Too thick dust collection will cause creeping and weaken the electric field, which is not conducive to spark generation. Discharge between the anode and the cathode produces sparks, and too low spark rate produces too few charged ions, which is not conducive to dust removal. However, too high spark rate and large current will weaken the electric field strength, and the anode plate will not have strong adhesion, which is also not conducive to dust removal. Therefore, the spark rate needs to be obtained, and the electric field parameters need to be controlled to achieve efficient dust removal.

[0073] Specifically, the DAU device refers to a data acquisition embedded system (Data Acquisition Unit Embedded System, DAU) device. The DAU device is connected to the switch. The plurality of electric fields can be divided into high-voltage side electric fields and low-voltage side electric fields, and the dust removal data can also include outlet dust amount, secondary current of each electric field on the high-voltage side, and rapping frequency of each electric field on the low-voltage side.

[0074] S103: Integrating the boiler data and the dust removal data according to a preset frequency, and generating access data;

[0075] It should be noted that the access data is the data after integration, and the access data is used as the input parameter of the subsequent model. After generating the access data, the access data can also be stored as historical data for subsequent analysis and use.

[0076] S104: Constructing a rapping parameter prediction model of the plurality of electric fields according to the access data;

[0077] Specifically, the steps of constructing the rapping parameter prediction model are as follows:

[0078] Step 1: Calculating the total dust amount of the boiler flue gas according to the boiler data;

[0079] The total dust amount of the boiler flue gas F can be calculated by the following formula:

[0080] The total dust amount of the boiler flue gas F = the amount of coal burned * the ash content of the coal entering the furnace * the amount of flue gas fly ash / (1 - the combustible content of the fly ash)

[0081] It should be noted that the boiler data can also include coal consumption, coal ash content, flue gas fly ash amount and fly ash combustibles, and in some other implementations of the present application, the total flue gas ash amount of the boiler can also be referred to as the total flue gas dust value or the total flue gas dust amount of the boiler.

[0082] Step two: calculate the dust removal efficiency according to the dust removal data;

[0083] The dust removal efficiency can be calculated by the following formula:

[0084]

[0085] The dust removal data can also include inlet dust concentration, outlet dust concentration, dust driving speed, dust removal dust accumulation area and flue gas amount. Specifically, B1 is the inlet dust concentration, B2 is the outlet dust concentration, S is the dust driving speed, D is the dust removal dust accumulation area, and Q is the flue gas amount.

[0086] Step three: calculate the theoretical dust amount of each electric field in the plurality of electric fields according to the total flue gas dust amount of the boiler and the dust removal efficiency;

[0087] Specifically, assuming that a total of five electric fields are required, the theoretical dust amount corresponding to the first electric field is FK, the theoretical dust amount corresponding to the second electric field is FK(1-K), the theoretical dust amount corresponding to the third electric field is FK(1-K) 2 , the theoretical dust amount corresponding to the fourth electric field is FK(1-K) 3 , and the theoretical dust amount corresponding to the fifth electric field is FK(1-K) 4 .

[0088] Step four: generate working condition data according to the total flue gas dust amount of the boiler, the dust removal efficiency and the theoretical dust value;

[0089] Step five: obtain a neural network model of the rapping parameters of the plurality of electric fields according to the access data and the working condition data.

[0090] S105: control the electric precipitator to rapping dust removal using the rapping parameters output by the rapping parameter prediction model.

[0091] In some implementations of the present application, the rapping parameter model can also be referred to as an optimization control model.

[0092] Specifically, the dust amount of each electric field is calculated after the two-way data including historical data such as boiler load, original flue gas flow, original flue gas temperature, original flue gas pressure, net flue gas flow, net flue gas temperature, net flue gas pressure, outlet dust amount, secondary current of each electric field on the high-voltage side, secondary voltage, spark rate, rapping frequency of each electric field on the low-voltage side, rapping period, and rapping time are acquired. Then, the change trend of the outlet dust is predicted according to the historical data such as the outlet dust amount, the boiler load, the energy consumption of each electric field, and the spark rate, a rapping parameter prediction model of each electric field under different change trends is constructed, the rapping parameters are continuously optimized according to the real-time value of the outlet dust amount and the spark rate feedback of each electric field, and the rapping control is performed according to the rapping parameters output by the rapping parameter prediction model.

[0093] Figure 1 According to the above-mentioned process, the boiler data and the dust removal data are integrated, and the access data is generated; the rapping parameter prediction model of the multiple electric fields is constructed according to the access data, and the closed-loop automatic control is realized through the model. Compared with the manual control method in the prior art, the operation difficulty of the operating personnel is reduced.

[0094] Please refer to Figure 2 The application also provides another closed-loop control method of the electric precipitator, which realizes more safe and stable automatic control of the rapping parameters through continuous optimization of the rapping parameters. Therefore, the process after Figure 1 The process after the process can further include a step of training the rapping parameter prediction model, and the embodiment of the application specifically includes the following steps:

[0095] S201: filtering the working condition data according to the correlation coefficient of the real value of the rapping parameters of the multiple electric fields to obtain working condition parameters;

[0096] It should be noted that the working condition parameters can include: inlet dust concentration, boiler load, flue gas pressure, flue gas temperature, energy consumption of multiple electric fields, theoretical dust amount, and dust concentration at the dust removal outlet;

[0097] Specifically, the correlation coefficient value of the working condition data and the rapping parameters can be calculated, and the working condition parameters with a higher correlation coefficient are selected as the input variable X according to a preset threshold value, and the output parameter Y is the rapping parameters of each electric field, such as rapping height, rapping time, rapping period, and rapping frequency. The correlation coefficient value of the working condition data and the rapping parameters can be calculated by the following calculation formula:

[0098]

[0099] where (X i , Y i ) is each working condition parameter and each rapping parameter sample point, X b and Y b are Xi with Y i the sample mean, n is the number of sample points, i is the loop variable in the summation symbol, and the value is 1 to n.

[0100] S202: Select the network depth parameter and the number of neurons of the neural network model according to the working condition parameters, and generate a rapping parameter prediction value;

[0101] The larger the data set of the general working condition parameters, the larger the network depth parameter of the neural network model and the more the number of neuron layers.

[0102] S203: Determine the loss function according to the rapping parameter true value and the rapping parameter prediction value.

[0103] Specifically, the loss function can be determined by the following calculation formula:

[0104] where Y di is the rapping parameter of the true historical data, is the rapping parameter prediction value output by the model.

[0105] S204: Train the rapping parameter prediction model according to the access data until the loss function value of the loss function tends to converge, and stop updating the working condition parameters.

[0106] Specifically, the rapping parameter prediction model is trained and optimized based on the access data until the loss function value tends to converge, and then the update of the working condition parameters is stopped; after the model training is completed, the values of each electric field rapping parameter can be calculated according to the current boiler load and dust and other working condition parameters. The specific training flowchart can be referred to Figure 5 .

[0107] In the embodiment of the application shown in Figure 2 , the rapping parameter prediction model is trained based on the loss function, and the rapping parameter is continuously optimized to realize more safe and stable rapping parameter automatic control.

[0108] Please refer to Figure 3 , the application also provides another closed-loop control method of the electric precipitator, which constructs a feedback compensation model through the spark rate, secondary current, secondary voltage and other parameters to feedback and compensate the rapping parameter in the prediction model, and realizes more safe and stable rapping parameter automatic control. Therefore, after the process of Figure 1 , the step of constructing the rapping parameter compensation model can be added, and the embodiment of the application specifically includes the following steps:

[0109] S301: Construct a rapping parameter compensation model of multiple electric fields;

[0110] It should be noted that when the spark rate of each of the multiple electric fields rises, the secondary current and voltage decrease, which means that the dust removal efficiency decreases, and measures such as reducing the vibration period and increasing the vibration height need to be taken. Specifically, the vibration parameter compensation model of the multiple electric fields can be constructed by using a data trend analysis method, based on the secondary current, secondary voltage, spark rate, and dust concentration at the dust removal outlet of each of the multiple electric fields, and the target set value of the dust at the dust removal outlet, to compensate for the vibration parameter prediction values of each electric field calculated by the vibration parameter prediction model constructed in S104. For the training process of a specific compensation model, please refer to Figure 6 .

[0111] In some implementations of the present application, the vibration parameter compensation model can also be referred to as an optimization compensation model, which is not described here.

[0112] S302: input the vibration parameter prediction value into the vibration parameter compensation model for compensation, and obtain the real value of the dust at the dust removal outlet;

[0113] After the vibration parameters of each electric field predicted by the vibration parameter prediction model are compensated by the vibration parameter compensation model, the control strategy is output to the dust removal control cabinet to obtain the real value of the dust at the dust removal outlet, and the real value of the dust at the dust removal outlet is fed back to the compensation model input to obtain error compensation by inputting the set value of the dust at the outlet, to realize closed-loop control and improve the control stability of the dust at the dust removal outlet.

[0114] S303: input the real value of the dust at the dust removal outlet and the preset set value of the dust into the vibration parameter compensation model to obtain an error compensation value.

[0115] Specifically, the preset set value of the dust refers to the set value of the dust at the same outlet as the real value of the dust at the dust removal outlet. The preset set value of the dust can be artificially preset or preset by a machine, which does not affect the implementation of the embodiment of the present application.

[0116] It should be noted that the error compensation value is obtained to realize closed-loop control and improve the stable control of the dust at the dust removal outlet.

[0117] S304: compensate the vibration parameter prediction value according to the error compensation value to obtain a compensated vibration parameter prediction value.

[0118] The compensated vibration parameter prediction value is obtained, and vibration is performed according to the compensated vibration parameter prediction value, and the vibration parameter prediction value is continuously compensated to achieve a more energy-saving vibration parameter.

[0119] In some implementations, S302 can be periodically executed.

[0120] The execution order of S302 and S303 can be performed in sequence or simultaneously.

[0121] In Figure 3 In the embodiment of the application shown, a feedback compensation model is constructed by using parameters such as the spark rate, the secondary current, and the secondary voltage, feedback compensation is performed on the knocking parameter of the feedforward prediction model, and more secure and more stable automatic control of the knocking parameter is realized.

[0122] As Figure 4 shown, the embodiment of the application also provides another closed-loop control method of the electric dust collector, in Figure 1 which the method is connected to the closed-loop control, and an alarm strategy can be set. The method specifically includes the following steps.

[0123] S401: When an abnormal condition occurs in the connected closed-loop control, alarm information is sent to the electric dust collector, and the alarm information is displayed.

[0124] It should be noted that the abnormal condition refers to an error in a certain link in the closed-loop control, which can refer to the connected data being greater than the value range that the electric dust collector can withstand, or the output knocking parameter not matching the actually operable parameter of the knocker. The alarm information can be in the form of flashing indicator lights, or in the form of a siren sound or a whistle sound emitted by an alarm device, and none of the above affects the implementation of the embodiment of the application.

[0125] Specifically, when the dust emission value at the dust removal outlet of the electric dust collector is greater than a first threshold value, first-level alarm is performed; when the dust emission value at the dust removal outlet of the electric dust collector is greater than a second threshold value, second-level alarm is performed; and when the dust emission value at the outlet of the chimney is greater than a third threshold value, alarm information is sent to the electric dust collector; wherein the priority of the first-level alarm is greater than the priority of the second-level alarm.

[0126] In some specific implementation modes of the application, the first threshold value can be 50 mg / Nm 3 , at which time first-level alarm is performed and automatic control is disconnected to switch to manual operation; the second threshold value can be 60 mg / Nm 3 , at which time second-level alarm is performed and automatic control is disconnected to switch to manual operation; and the third threshold value can be 3 mg / Nm 3 , at which time alarm is performed and automatic control is disconnected to switch to manual operation. It should be noted that the values involved in the embodiment of the application can also be other arbitrary limited values, and none of the above affects the implementation of the embodiment of the application.

[0127] S402: According to the alarm information, the current working mode of the electric dust collector is switched to a manual operation mode.

[0128] It should be noted that the current working mode can be an automatic control mode, or other working modes, which do not affect the implementation of the embodiment of the application.

[0129] In Figure 4 In the embodiment of the application shown, by setting the alarm strategy, the situation of excessive export discharge can be reduced.

[0130] As Figure 7 The embodiment of the application also provides a closed-loop control device of an electric dust collector, which specifically comprises: a data interaction system 701, an intelligent control server 702, a communication card 703, a distributed control system (DCS) control cabinet 704, a data acquisition embedded system (DAU) device 705, and an electric dust collector 706:

[0131] The data interaction system 701 is configured to acquire boiler data from the DCS control cabinet 704 through the communication card 703.

[0132] The data interaction system 701 is also configured to acquire dust removal data from the DAU device 705, wherein the dust removal data comprises electric field parameters and rapping parameters of a plurality of electric fields.

[0133] The intelligent control server 702 is configured to integrate the boiler data and the dust removal data according to a preset frequency, and generate access data.

[0134] The intelligent control server 702 is also configured to construct a rapping parameter prediction model of the plurality of electric fields according to the access data.

[0135] The intelligent control server 702 is also configured to control the electric dust collector 706 to perform rapping dust removal by using the rapping parameters output by the rapping parameter prediction model.

[0136] In some implementations of the application, the intelligent control server 702 can also be referred to as an intelligent server, which is not described herein.

[0137] It should be noted that the closed-loop control device can also include a monitoring station, a desulfurization device, and the like, and does not limit the implementation of the embodiment of the application.

[0138] As Figure 8 The embodiment of the application also provides another closed-loop control device of an electric dust collector, which specifically comprises:

[0139] The acquisition module 801 is configured to acquire boiler data from a DCS control cabinet through a communication card.

[0140] The acquisition module 802 is configured to acquire dust removal data from a DAU device, wherein the dust removal data comprises electric field parameters and rapping parameters of a plurality of electric fields.

[0141] The generating module 803 is configured to integrate the boiler data and the dust removal data according to a preset frequency, and generate access data.

[0142] The model constructing module 804 is configured to construct a vibration parameter prediction model of multiple electric fields according to the access data.

[0143] The vibration module 805 is configured to control the electric dust collector to vibrate and remove dust by using the vibration parameters output by the vibration parameter prediction model.

[0144] It can be understood that the structure illustrated in the embodiment does not constitute a specific limitation on the device. In other embodiments, the device can include more or fewer components than the illustration, or combine other components, or split other components, or different component arrangements. The illustrated components can be implemented in hardware, software, or a combination of software and hardware.

[0145] As shown in the Figure 9 embodiments, the present application also provides a computer device, which comprises a memory 901 and a processor 902.

[0146] The memory 901 is configured to store a program.

[0147] The processor 902 is configured to execute the program in the memory to implement the above-mentioned electric dust collector closed-loop control method. Figures 1 to 4

[0148] Finally, it also needs to be explained that in the embodiments of the present application, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0149] ​The foregoing description of the disclosed embodiments enables a person skilled in the art to make or use the application. Modifications of these embodiments will occur to those skilled in the art and are within the scope of the following claims, defined solely by the words of the claims themselves, in which the patent is sought. Therefore, the application is not limited to the described embodiments and instead is intended to cover any and all modifications within the scope of the following claims.

Claims

1. A closed-loop control method for an electrostatic precipitator, characterized in that, The method includes: Boiler data is acquired from the DCS control cabinet of the distributed control system via a communication card, which includes an IO communication card and performs data transfer between the distributed control system and the acquisition server. Dust removal data is acquired from the data acquisition embedded system DAU device, and the dust removal data includes electric field parameters and rapping parameters of multiple electric fields; The boiler data and dust removal data are integrated according to a preset frequency, and access data is generated. Based on the access data, a rapping parameter prediction model for the multiple electric fields is constructed; The step of constructing the rapping parameter prediction model for the multiple electric fields based on the access data includes: calculating the total dust amount in the boiler flue gas based on the boiler data and the calculation formula for the total dust amount in the boiler flue gas, wherein the calculation formula for the total dust amount in the boiler flue gas includes: Total ash content in boiler flue gas = Coal consumption * Ash content of coal fed into the furnace * Fly ash content in flue gas / (1 - Combustible matter in fly ash); The dust removal efficiency is calculated based on the dust removal data and the dust removal efficiency calculation formula, wherein the dust removal efficiency calculation formula includes: ； Where K is the dust removal efficiency, B1 is the inlet dust concentration, B2 is the outlet dust concentration, S is the dust driving velocity, D is the dust collection area, and Q is the flue gas volume. The theoretical dust amount of each of the multiple electric fields is calculated based on the total dust amount of the boiler flue gas and the dust removal efficiency; operating condition data is generated based on the total dust amount of the boiler flue gas, the dust removal efficiency, and the theoretical dust value; and a neural network model of the rapping parameters of the multiple electric fields is obtained based on the access data and the operating condition data. The rapping parameters output by the rapping parameter prediction model are used to control the electrostatic precipitator for rapping dust removal. Constructing a rapping parameter compensation model for the multiple electric fields; wherein, the method for constructing the rapping parameter compensation model may include: using a data trend analysis method, based on the secondary current, secondary voltage, spark rate, dust concentration at the dust removal outlet of each electric field in the multiple electric fields, and the target set value of dust at the dust removal outlet, to construct the rapping parameter compensation model; The predicted values ​​of the rapping parameters are input into the rapping parameter compensation model for compensation, and the true value of dust is obtained. The true value of dust and the preset dust setting value are input into the rapping parameter compensation model to obtain the error compensation value. The predicted values ​​of the rapping parameters are compensated according to the error compensation value to obtain the compensated predicted values ​​of the rapping parameters.

2. The method according to claim 1, characterized in that, The method further includes: The operating data is filtered based on the correlation coefficient of the true values ​​of the rapping parameters of the multiple electric fields to obtain operating parameters, which include: inlet dust concentration, boiler load, flue gas pressure, flue gas temperature, energy consumption of the multiple electric fields, theoretical dust amount, and dust concentration at the dust removal outlet. Based on the operating conditions, the network depth parameters and the number of neurons of the neural network model are selected, and the predicted values ​​of the vibration parameters are generated. The loss function is determined based on the actual values ​​of the vibration parameters and the predicted values ​​of the vibration parameters. The vibration parameter prediction model is trained based on the access data until the loss function value of the loss function tends to converge, and then the updating of the operating condition parameters is stopped.

3. The method according to claim 1, characterized in that, The method further includes: When an abnormal situation occurs in the closed-loop control, an alarm message is sent to the electrostatic precipitator and the alarm message is displayed.

4. The method according to claim 3, characterized in that, When an abnormal situation occurs in the closed-loop control, an alarm message is sent to the electrostatic precipitator, including: When the dust emission value at the dust removal outlet of the electrostatic precipitator exceeds the first threshold, a level one alarm is triggered. When the dust emission value at the dust removal outlet of the electrostatic precipitator exceeds the second threshold, a level two alarm is triggered. When the dust emission value at the chimney outlet exceeds the third threshold, the alarm information is sent to the electrostatic precipitator. The priority of the first-level alarm is higher than that of the second-level alarm.

5. The method according to claim 3 or 4, characterized in that, The method further includes: Based on the alarm information, the current operating mode of the electrostatic precipitator is switched to manual operation mode.

6. A closed-loop control device for an electrostatic precipitator, characterized in that, The closed-loop control device includes a data interaction system, an intelligent control server, communication cards, and an electrostatic precipitator. The data interaction system is used to acquire boiler data from the DCS control cabinet of the distributed control system through the communication card. The communication card includes an IO communication card, which performs data transfer between the distributed control system and the acquisition server. The data interaction system is also used to acquire dust removal data from the data acquisition embedded system DAU device, the dust removal data including electric field parameters and rapping parameters of multiple electric fields; The intelligent control server is used to integrate the boiler data and the dust removal data according to a preset frequency, and generate access data; The intelligent control server is also used to construct a rapping parameter prediction model for the multiple electric fields based on the access data; The intelligent control server constructs a prediction model for the rapping parameters of the multiple electric fields based on the access data, specifically for: calculating the total amount of dust in the boiler flue gas according to the boiler data and the calculation formula for the total amount of dust in the boiler flue gas, wherein the calculation formula for the total amount of dust in the boiler flue gas includes: Total ash content in boiler flue gas = Coal consumption * Ash content of coal fed into the furnace * Fly ash content in flue gas / (1 - Combustible matter in fly ash); The dust removal efficiency is calculated based on the dust removal data and the dust removal efficiency calculation formula, wherein the dust removal efficiency calculation formula includes: ； Where K is the dust removal efficiency, B1 is the inlet dust concentration, B2 is the outlet dust concentration, S is the dust driving velocity, D is the dust collection area, and Q is the flue gas volume. The theoretical dust amount of each of the multiple electric fields is calculated based on the total dust amount of the boiler flue gas and the dust removal efficiency; operating condition data is generated based on the total dust amount of the boiler flue gas, the dust removal efficiency, and the theoretical dust value; and a neural network model of the rapping parameters of the multiple electric fields is obtained based on the access data and the operating condition data. The intelligent control server is also used to control the electrostatic precipitator to perform rapping dust removal using the rapping parameters output by the rapping parameter prediction model; The intelligent control server is further configured to use a data trend analysis method to construct the rapping parameter compensation model based on the secondary current, secondary voltage, spark rate, and dust concentration at the dust removal outlet of each of the multiple electric fields, and the target set value of the dust removal outlet; input the predicted value of the rapping parameters into the rapping parameter compensation model for compensation, and obtain the true value of the dust removal; input the true value of the dust removal and the preset dust set value into the rapping parameter compensation model to obtain the error compensation value; and compensate the predicted value of the rapping parameters according to the error compensation value to obtain the compensated predicted value of the rapping parameters.

7. A closed-loop control device for an electrostatic precipitator, characterized in that, The closed-loop control device includes: The acquisition module is used to acquire boiler data from the DCS control cabinet of the distributed control system via a communication card. The communication card includes an IO communication card, which performs data transfer between the distributed control system and the acquisition server. The acquisition module is used to acquire dust removal data from the data acquisition embedded system DAU device. The dust removal data includes electric field parameters and rapping parameters of multiple electric fields. The generation module is used to integrate the boiler data and the dust removal data according to a preset frequency, and generate access data; The model building module is used to build a rapping parameter prediction model for the multiple electric fields based on the access data; Specifically, the model building module is used to: calculate the total dust content of boiler flue gas based on the boiler data and the calculation formula for the total dust content of boiler flue gas, wherein the calculation formula for the total dust content of boiler flue gas includes: Total ash content in boiler flue gas = Coal consumption * Ash content of coal fed into the furnace * Fly ash content in flue gas / (1 - Combustible matter in fly ash); The dust removal efficiency is calculated based on the dust removal data and the dust removal efficiency calculation formula, wherein the dust removal efficiency calculation formula includes: ； Where K is the dust removal efficiency, B1 is the inlet dust concentration, B2 is the outlet dust concentration, S is the dust driving velocity, D is the dust collection area, and Q is the flue gas volume. The theoretical dust amount of each of the multiple electric fields is calculated based on the total dust amount of the boiler flue gas and the dust removal efficiency; operating condition data is generated based on the total dust amount of the boiler flue gas, the dust removal efficiency, and the theoretical dust value; and a neural network model of the rapping parameters of the multiple electric fields is obtained based on the access data and the operating condition data. The rapping module is used to control the electrostatic precipitator to perform rapping dust removal using the rapping parameters output by the rapping parameter prediction model; The model building module is further configured to use a data trend analysis method to construct the rapping parameter compensation model based on the secondary current, secondary voltage, spark rate, and dust concentration at the dust removal outlet of each of the multiple electric fields, and the target set value of the dust removal outlet; input the predicted value of the rapping parameters into the rapping parameter compensation model for compensation, and obtain the true value of the dust removal; input the true value of the dust removal and the preset dust set value into the rapping parameter compensation model to obtain the error compensation value; and compensate the predicted value of the rapping parameters according to the error compensation value to obtain the compensated predicted value of the rapping parameters.

8. A computer device, characterized in that, The device includes a memory and a processor, the processor being configured to execute a program stored in the memory, performing the method as described in any one of claims 1-5.

Images