Burner adjusting control system based on artificial intelligence

Abstract

The invention discloses a burner regulation control system based on artificial intelligence, which relates to the field of burner regulation, and is characterized in that a data layer and a processing layer are designed based on NET open source architecture, and a database comprises a cache of processing layer related information, burner equipment information and user management information; distributed storage and calling management are carried out; the processing layer comprises a monitoring module and a judgment module, the monitoring module obtains the combustion index of the mixed gas through an evaluation formula and provides data support for judgment of the judgment module, and the judgment module generates different regulation and control instructions according to judgment results; the regulation and control layer comprises a PID-speed regulation and control model and a neural network model, a proportionality coefficient, an integral coefficient and a differential coefficient of the PID-speed regulation and control model are adaptively adjusted through the neural network model, and the display layer is used for inputting and outputting an application instruction and realizing adaptive adjustment of the gas transmission speed.

Description

Technical Field

[0001] The present invention relates to the field of burner regulation, and more specifically, to a burner regulation control system based on artificial intelligence. Background Art

[0002] After the gas enters the burner, it mixes with the compressed air and burns to produce high-temperature and high-pressure flue gas. The flue gas drives the turbine at the rear to expand and rotate, dragging the generator at the rear to generate electricity or drive other industrial equipment. The burner is generally installed on the front end cover of the combustion chamber and is distributed in a circle. Failure of any burner may cause combustion fluctuations and may cause shutdown accidents. In severe cases, it may cause thermal acoustic oscillations and damage burner components. If there is a problem with the distribution of burner gas flow, the combustion flame will be deflected, and the local temperature of the flue gas will be too high. In severe cases, the burner will be burned and the rear turbine blades will be endangered, causing blade burning. The burner has multiple fuel nozzles, and the various fuels cooperate with each other during the operation of the unit to maintain a high combustion efficiency. During the startup or operation of the unit, if the fuel ratio of individual burners deviates from the design value, such as abnormal ratio of duty fuel and premixed fuel, it will cause abnormal combustion temperature and pressure. The reduction in fuel flow will cause the temperature to be too low, which will lead to combustion instability or unsuccessful combustion. Especially for the duty nozzle, if the fuel is abnormally reduced and the air flow used for combustion assistance remains unchanged, it will cause the combustion temperature to be too low and the flame stability to deteriorate. When combustion is not required, the abnormal inlet flow will not only cause unsuccessful combustion, but also may cause incomplete combustion. If the combustion of the surrounding premixed nozzles is abnormal, it will cause combustion instability, cause violent fluctuations in combustion, and even cause thermal acoustic oscillation events. At the same time, abnormal flow will also cause the combustion temperature to rise, which may cause the combustion temperature to be too high, excessive nitrogen oxides to be generated, and pollutant emissions to exceed the standard. Therefore, during the combustion process, the air mixing flow needs to be regulated. Manual regulation is not only time-consuming and labor-intensive, but also has flow accuracy control deviations, so the flow needs to be regulated intelligently. Summary of the invention

[0003] In view of the problems existing in the prior art in the burner flow control working process, the purpose of the present invention is to provide a burner regulation and control system based on artificial intelligence to achieve stability and rapidity of flow control during the burner operation process.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A burner regulation control system based on artificial intelligence, the air mixing regulation system comprises a data layer, a processing layer, a regulation layer and a display layer;

[0006] The data layer includes a database and a call group. The data layer and the processing layer are designed based on the NET open source architecture. The database includes a cache of information related to the processing layer and the device information of the burner, as well as user management information, and performs distributed storage and call management.

[0007] The processing layer includes a monitoring module and a judgment module. The monitoring module obtains the combustion index of the mixed gas through an evaluation formula to provide data support for the judgment of the judgment module. The judgment module generates different control instructions according to the judgment results.

[0008] The control layer includes a PID-speed control model and a neural network model. The proportional coefficient K of the PID-speed control model is adjusted by the neural network model. B , integral coefficient K J , differential coefficient K W Adaptively adjust the input speed of a single gas in the mixed gas to the desired transmission speed;

[0009] The display layer is used for input and output of application instructions and for terminal page display. The application instructions of the display layer include production instructions.

[0010] Preferably, the user management information includes user identity authentication and authority management. User identity authentication is applied to user login. The calling group checks the identity information input by the user when logging in with the identity information stored in the database to determine whether the user has login authority. Authority management designs roles through a authority model, assigns corresponding roles according to the user's authenticated identity, and obtains corresponding authorities based on the role.

[0011] Through the distributed design of the database, not only is it convenient to respond quickly when reading data, but it also realizes the elastic expansion of storage space and improves the operating performance of the entire system; through the establishment of roles and permissions to manage user information, users are given different roles and obtain different permissions, thus achieving the strictness of users in the system management process and the efficiency of data information management.

[0012] Preferably, the monitoring module is connected to a monitoring device, and the monitoring module processes the flow values ​​on each feed pipeline obtained by the monitoring device. The monitoring device includes a flow measurement device and a pressure measurement device. The fuel state of the burner at the current moment is obtained according to the numerical processing of the monitoring module. The specific working process of the monitoring module includes the following steps:

[0013] Step S31: The pressure measuring device transmits the pressure of the gas in each feed pipeline to the monitoring module through the wireless network, and the flow measuring device transmits the flow of the gas in each feed pipeline to the monitoring module through the wireless network, obtains the flow of a single gas in the feed pipeline and the flow of the mixed gas, and obtains the combustion index of the mixed gas according to the evaluation formula:

[0014] Formula (1)

[0015] Formula (1) is used to obtain the combustion index Rr of the burner after mixing a single gas;

[0016] Where μ represents the turbulence intensity coefficient in the mixed gas, P i represents the pressure of a single gas with label i, Pe represents the overall pressure of the mixed gas after all the single gases are mixed, m i represents the mass of a single gas with label i, s represents the cycle time set for a measurement cycle, and q i Indicates the flow rate of a single gas with label i, λ i represents the concentration ratio of the single gas with label i, T represents the average temperature of the mixed gas after all the single gases are mixed, ε represents the ratio of the average temperature, i represents the label of any single gas in the mixed gas, and I represents the number of all the single gases in the mixed gas;

[0017] Step S32: the monitoring module transmits the combustion index Rr of the mixed single gases to the judgment module.

[0018] Preferably, the judgment module is used to judge the combustion degree of the mixed gas, output different control instructions according to the judgment result, and perform feedback control on the production instructions issued in the display layer. The specific working process includes the following steps:

[0019] Step S41: the judgment module receives the combustion index Rr of the mixed single gases in the monitoring module;

[0020] Step S42: If Rr≥a, it means that the combustion index Rr of the mixed gas in the current measurement cycle is greater than or equal to the combustion critical value a, that is, the combustion index is not within the normal value range, and the mixed gas meets the combustion standard;

[0021] If the instruction input by the user in the display layer is a combustion production instruction, the mixed gas in the current measurement cycle has reached the combustion standard, and the burner can be burned by ignition; if the instruction input by the user in the display layer is a non-combustion production instruction, it is necessary to control the mixed gas that has reached the combustion standard in the current measurement cycle, and the monitoring module issues a control command to the control layer;

[0022] Step S43: if Rr<a, it means that the combustion index Rr of the mixed gas in the current measurement cycle is less than the combustion critical value a, that is, the combustion index is within the normal value range, and the mixed gas is normal and does not meet the combustion standard;

[0023] If the instruction input by the user in the display layer is a combustion production instruction, the mixed gas in the current measurement cycle does not meet the combustion standard, and it is necessary to regulate the individual gas in the mixed gas, and the monitoring module issues a regulation command to the regulation layer; if the instruction input by the user in the display layer is a non-combustion production instruction, the mixed gas in the current measurement cycle needs to meet the user's production standard, and no additional regulation is required.

[0024] The combustion index of the mixed gas is obtained according to the evaluation formula to provide data support for the state judgment of the mixed gas in the judgment module. The judgment module generates different application instructions according to the judgment results to realize the analysis and judgment of the burner under different mixed gas composition states, providing detailed and reliable data support for the regulation and management of the control layer.

[0025] Preferably, the specific working process of the monitoring module and the judgment module in the processing layer includes the following steps:

[0026] Step S51: Each feed pipe of the burner is numbered, and the measurement cycle and cycle time are set. The flow measurement device stably obtains the flow value of the gas in each feed pipe, and the pressure measurement device stably obtains the pressure value of the gas in each feed pipe, and stably obtains the mixed gas pressure of the single gas in the mixed feed pipe. The user sets each parameter in the evaluation formula through the display layer;

[0027] Step S52: the monitoring module brings the values ​​transmitted by the pressure measuring device and the flow measuring device into the evaluation formula to obtain the combustion index of the burner after the single gas is mixed;

[0028] Step S53: The judgment module judges whether the combustion index of the mixed gas reaches the ignition combustion standard based on the production instructions entered by the user from the display layer, and outputs different control instructions to the control layer according to the judgment result. The control layer controls the flow input of a single gas according to the requirements.

[0029] Preferably, the control layer controls the input speed of a single gas based on the PID control of the neural network, and then feedback-adjusts the flow rate of the single gas according to the control of the input speed, so that the concentration of the single gas in the mixed gas reaches the relevant standard. The control layer constructs a PID-speed control model, and the specific construction process of the PID-speed control model is as follows:

[0030] Step S61: Continuously discretize the cycle time s of a measurement cycle into multiple unit times t to construct a discrete control model.

[0031] (2)sv(s)=de(n)[v(t)+e(t)]

[0032] (3) v(t) = v(t-1) + Δv(t)

[0033] (4)Δv(t)=K B [e(t)-e(t-1)]+K J e(t)+K W [e(t)+e(t-2)-2e(t-1)] Formula (2) is used to show that the gas input speed output in each iteration approaches the preset input speed;

[0034] Formula (3) is used to show the incremental control expression of iterative gas input speed;

[0035] Formula (4) is used to show the expression for obtaining Δv(t);

[0036] Wherein, sv(s) represents the expected transmission speed set when a single gas reaches the combustion concentration within the periodic measurement time s, v(t) represents the gas input speed output after each iteration, which is simplified as the iterative output speed, e(t) represents the absolute value of the deviation between the iterative output speed v(t) per unit time t and the expected transmission speed sv(s), v(t-1) represents the transmission speed value actually output at the previous moment of the current sampling moment t, Δv(t) represents the actual transmission speed difference between the current sampling moment and the previous sampling moment, e(t-1) represents the absolute value of the deviation within the previous sampling moment of unit time t, e(t-2) represents the absolute value of the deviation within the previous sampling moment of unit time t-1, de() represents the iterative function; n represents the number of iterations; K B is the proportionality coefficient; K J Indicates the integral coefficient, K W represents the differential coefficient.

[0037] Step S62: Based on the gas input speed output after each iteration, the proportionality coefficient K B , integral coefficient K J , differential coefficient K W Adaptive adjustment is performed, and finally the gas input speed actually output by the PID-speed control model is continuously approached to the expected transmission speed sv(s).

[0038] Preferably, the control layer is based on the proportional coefficient K in the PID control of the neural network B , integral coefficient K J , differential coefficient K W Adaptive correction and adjustment are performed. The neural network consists of an input layer, a hidden layer, and an output layer. The proportional coefficient K is updated by updating the weights. B , integral coefficient K J , differential coefficient K W The specific process of adjusting and building the neural network model is as follows:

[0039] (5)

[0040] (6)

[0041] (7) Formula (5) is used to express the relationship between the input of the given input layer and the input and output of the hidden layer;

[0042] Among them, b represents the current iteration number, m() represents the activation function, It indicates the input of the given input layer with label j, o (1) represents the input of a given input layer, represents the input of the hidden layer labeled l, c (2) represents the input of the hidden layer, represents the output of the hidden layer labeled h, O (2) represents the output of the hidden layer, Represents the weight coefficient from the input layer to the hidden layer;

[0043] In addition, the number of input layers is set to M, and j is the number of any one of the M input layers, the number of input layers in the hidden layer is set to L, and l is the number of any one of the L hidden layers, the number of output layers in the hidden layer is set to H, and h is the number of any one of the H hidden layers, the number of output layers is set to 3, and k is the number of any one of the 3 output layers;

[0044] Formula (6) is used to show the input and output expressions of the output layer;

[0045] Among them, g() represents the activation function, represents the weight coefficient from the hidden layer to the output layer, Represents the output variable, corresponding to the three parameters controlled by PID. represents the input variable of the output layer, represents the input of the hidden layer labeled l, It indicates the input variable of the output layer when the label h takes the value of 1, 2, or 3;

[0046] Formula (7) is used for weighting coefficient and The update expression of , α is the inertia coefficient, η is the learning efficiency, is the intermediate variable, For the bth iteration and the b-1th iteration The weighted coefficient difference of For the bth iteration and the b-1th iteration The difference in weighting coefficients of and.

[0047] Preferably, the control process of the regulation layer for PID control based on the neural network includes the following steps:

[0048] Step S81: Determine the neural network structure, set the number of layers of the input layer, hidden layer and output layer respectively. The number of layers of the output layer in the neural network model constructed this time is 3, and set the proportional coefficient K of PID according to the PID-speed control model. B , integral coefficient K J , differential coefficient K W And other parameters to determine the initial value, select the appropriate inertia coefficient and learning efficiency;

[0049] Step S82: The flow measurement device collects the gas input speed output per unit time t, inputs the expected transmission speed value during the bobbin working process, calculates and obtains the absolute value of the deviation between the iterative output speed v(t-1) per unit time t-1 and the expected transmission speed sv(s), and calculates and obtains v(t) at the unit time t=1;

[0050] Step S83: Calculate the input and output of each node in each layer of the neural network through the neural network model to obtain the proportional coefficient K of the modified PID control B , integral coefficient K J , differential coefficient K W ;

[0051] Step S84: Calculate v(t+1) when t=t+1, and repeat steps S82 and S84.

[0052] By constructing a PID-speed control model, the actual transmission speed and the expected transmission speed are jointly expressed by formulas. By accessing the neural network model, the number of input layers, hidden layers and output layers is set based on the adaptive update of the weighted coefficients in the neural network model, thereby realizing automatic control of the input speed of a single gas by the PID-speed control model based on the neural network, improving the control performance of the actual transmission speed approaching the expected transmission speed, and improving the speed, accuracy and stability in the speed approximation process.

[0053] Preferably, the technical solution of the air mixing and conditioning system mainly includes the following steps:

[0054] Step S1: The user performs identity verification through the client's Web browsing page to determine whether to allow login to the system, and obtains corresponding access and management permissions according to the role assigned by the user identity. The user manages the gas parameters and control parameters of the burner by carrying the role;

[0055] Step S2: The burner is continuously monitored through the pressure measuring device and the flow measuring device. The monitoring module of the processing layer obtains the combustion index of the mixed gas through the evaluation formula. The judgment module generates different application instructions according to the judgment result. The control layer performs relevant processing according to the signal information sent by the application instruction.

[0056] Step S3: The control layer establishes an expression for the gas input speed through the PID-speed control model, and connects the PID-speed control model to the neural network model to adjust the proportional coefficient K of the PID control. B , integral coefficient K J , differential coefficient K W Make adaptive adjustments;

[0057] Step S4: Repeat steps S2 to S4.

[0058] Preferably, the air mixing and conditioning system is applied to a cloud processing platform, the data layer is applied to the cloud, the central processing unit is arranged on the cloud processing platform and adopts an integrated information processing system, the processing layer and the control layer perform cloud processing based on the central processing unit, the central processing unit is connected to the monitoring equipment, the burner and the client through a wireless network or wired transmission, the display layer displays information through the web page program of the output device, outputs the processing results of the central processing unit, and provides information for management personnel to receive, the output layer adopts a multi-client information architecture, users access the system through the client, and have permission to access information according to the assigned role.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] 1. In the present invention, by distributing the database, it is not only convenient to respond quickly when reading data, but also to realize the elastic expansion of storage space, thereby improving the operation performance of the whole system; by establishing roles and permissions to manage and design user information, different roles are given to users to obtain different permissions, thereby realizing the rigor of the user in the system management process and the management efficiency of data information, and obtaining the combustion index of the mixed gas according to the evaluation formula to provide data support for the state determination of the mixed gas in the judgment module. The judgment module generates different application instructions according to the judgment result, thereby realizing the analysis and determination of the burner under different mixed gas composition states, and providing detailed and reliable data support for the regulation and management of the control layer.

[0061] 2. In the present invention, by constructing a PID-speed control model, the actual transmission speed and the expected transmission speed are jointly expressed by formulas, and by accessing a neural network model, the number of layers of each input layer, hidden layer and output layer is set based on the adaptive update of the weighted coefficients in the neural network model, thereby realizing automatic control of the input speed of a single gas by the PID-speed control model based on the neural network, improving the control performance of the actual transmission speed approaching the expected transmission speed, and improving the speed, accuracy and stability in the speed approximation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 The present invention proposes a structural schematic diagram of a burner regulation control system based on artificial intelligence;

[0063] Figure 2 It is a schematic diagram of the method steps for treating a layer in the present invention;

[0064] Figure 3 This is a schematic diagram of the method steps for regulating the layer in the present invention;

[0065] Figure 4 It is a schematic diagram of the method steps of the air mixing and conditioning system of the present invention;

[0066] Figure 5 is a schematic diagram of the neural network structure in the present invention;

[0067] Figure 6 The diagram shows an application scenario of the present invention. DETAILED DESCRIPTION

[0068] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0069] Embodiment 1

[0070] The burners are generally installed on the front end cover of the combustion chamber and are distributed in a circular pattern. Failure of any burner may cause combustion fluctuations and may cause shutdown accidents. In severe cases, it may cause thermal acoustic oscillations and damage burner components. If there is a problem with the distribution of burner gas flow, the combustion flame will be deflected, and the local temperature of the flue gas will be too high. In severe cases, the burner will be burned and the rear turbine blades will be endangered, causing blade burning. The burner has multiple fuel nozzles, and the various fuels cooperate with each other during the operation of the unit to maintain a high combustion efficiency. During the startup or operation of the unit, if the fuel ratio of individual burners deviates from the design value, such as abnormal ratio of duty fuel and premixed fuel, it will cause abnormal combustion temperature and pressure. The reduction in fuel flow will cause the temperature to be too low, which will lead to combustion instability or unsuccessful combustion. Especially for the duty nozzle, if the fuel is abnormally reduced and the air flow used for combustion assistance remains unchanged, it will cause the combustion temperature to be too low and the flame stability to deteriorate. When combustion is not required, the abnormal inlet flow will not only cause unsuccessful combustion, but also may cause incomplete combustion. If the combustion of the surrounding premixed nozzles is abnormal, it will cause combustion instability, cause violent fluctuations in combustion, and even cause thermal acoustic oscillation events. At the same time, abnormal flow will also cause the combustion temperature to rise, which may cause the combustion temperature to be too high, excessive nitrogen oxides to be generated, and pollutant emissions to exceed the standard. Therefore, during the combustion process, the air mixing flow needs to be regulated. Manual regulation is not only time-consuming and labor-intensive, but also has flow accuracy control deviations, so the flow needs to be regulated intelligently.

[0071] Reference Figure 1 , Figure 2 , Figure 4 and Figure 6 , Example 1 further illustrates a burner regulation and control system based on artificial intelligence proposed by the present invention.

[0072] A burner regulation and control system based on artificial intelligence, the air mixing regulation system includes a data layer, a processing layer and a display layer; the data layer includes a database and a call group, the data layer and the processing layer are designed based on the NET open source architecture, the database includes a cache of relevant information of the processing layer and equipment information of the burner, as well as user management information, and performs distributed storage and call management; the display layer is used for the input and output of application instructions, and performs end page display, and the application instructions of the display layer include production instructions.

[0073] The display layer includes a display device, which is connected to the central processor through a wired interface or a wireless network. The user inputs commands or instructions through the page end on the display device, such as initial setting of various parameters in formula (1) and parameters in other formulas, or issuing control commands for gas flow in a pre-mixed state of gas during ignition or non-combustion, and issuing control commands for the focused input of a single gas during the combustion process.

[0074] The user management information includes user identity authentication and authority management. User identity authentication is applied to user login. The calling group verifies the identity information entered by the user when logging in with the identity information stored in the database to determine whether the user has login authority. Authority management designs roles through a authority model, assigns corresponding roles according to the user's authenticated identity, and obtains corresponding authorities based on the role.

[0075] User management information includes user identity authentication and permission management. User identity authentication is applied to user login. The calling group verifies the identity information entered by the user when logging in with the identity information stored in the database to determine whether the user has login permission. Permission management designs roles through the permission model, assigns corresponding roles according to the user's authentication identity, and obtains corresponding permissions according to the role. Users include administrators and operators. Administrators issue commands through the display layer, as well as other production information planning and entry. Operators are used for equipment maintenance and material addition to ensure timely operation of the machine. The server stores user information in the database. When the user performs permission authentication, the calling group retrieves the current user's information and the role he has from the database, queries the corresponding permissions associated with the role, determines whether the user has the permission to operate a certain resource, and completes permission authentication.

[0076] The present invention adopts the role-based access control permission model for design. The permission model introduces the concept of role, explains the relationship between permission and user, and assigns permission to role rather than user. Certain roles are assigned according to user responsibilities, and users obtain corresponding permissions according to roles. By associating users with roles, and roles with resources or operations, granting users roles means granting users the permissions that the roles have.

[0077] When the user logs into the system, the central processing unit will store the user information in the database. When user authority authentication is required, the current user information and the role will be retrieved from the database, the corresponding authority associated with the role will be queried, and it will be determined whether the user has the authority to operate the corresponding level of power security data group to complete the authority authentication. The authority authentication is completed using the filter. The authority filter will intercept each request sent by the client. When the request is intercepted, the domain will be searched for the operating authority of the currently logged-in user to determine whether the request can operate the resource. If the authority is not available, an error message will be returned.

[0078] The processing layer includes a monitoring module and a judgment module. The monitoring module obtains the combustion index of the mixed gas through an evaluation formula to provide data support for the judgment of the judgment module. The judgment module generates different control instructions according to the judgment results.

[0079] The control layer includes a PID-speed control model and a neural network model. The proportional coefficient K of the PID-speed control model is adjusted by the neural network model. B , integral coefficient K J , differential coefficient K W Adaptively adjust the input speed of a single gas in the mixed gas to the desired transmission speed;

[0080] The monitoring module is connected to the monitoring device. The monitoring module processes the flow values ​​on each feed pipeline obtained by the monitoring device. The monitoring device includes a flow measurement device and a pressure measurement device. The fuel state of the burner at the current moment is obtained according to the numerical processing of the monitoring module. The specific working process of the monitoring module includes the following steps:

[0081] Step S31: The pressure measuring device transmits the pressure of the gas in each feed pipeline to the monitoring module through the wireless network, and the flow measuring device transmits the flow of the gas in each feed pipeline to the monitoring module through the wireless network, obtains the flow of a single gas in the feed pipeline and the flow of the mixed gas, and obtains the combustion index of the mixed gas according to the evaluation formula:

[0082] Formula (1)

[0083] Formula (1) is used to obtain the combustion index Rr of the burner after mixing a single gas;

[0084] Where μ represents the turbulence intensity coefficient in the mixed gas, P i represents the pressure of a single gas with label i, Pe represents the overall pressure of the mixed gas after all the single gases are mixed, m i represents the mass of a single gas with label i, s represents the cycle time set for a measurement cycle, and q i Indicates the flow rate of a single gas with label i, λ i represents the concentration ratio of the single gas with label i, T represents the average temperature of the mixed gas after all the single gases are mixed, ε represents the ratio of the average temperature, i represents the label of any single gas in the mixed gas, and I represents the number of all the single gases in the mixed gas;

[0085] Step S32: the monitoring module transmits the combustion index Rr of the mixed single gases to the judgment module.

[0086] In formula (1) It is used to indicate the friction between individual gases in a mixed gas. It is used to determine the basis for the combustion reaction through the interaction between the molecules in the gas. iThe input weight of a single gas in the feed pipeline within a measurement cycle can be measured by a weight weighing device. T is the real-time temperature of the mixed gas when it is not ignited, that is, the temperature of the mixed gas without an external fire source.

[0087] The judgment module is used to judge the combustion degree of the mixed gas, output different control instructions according to the judgment result, and perform feedback control on the production instructions issued in the display layer. The specific working process includes the following steps:

[0088] Step S41: the judgment module receives the combustion index Rr of the mixed single gases in the monitoring module;

[0089] Step S42: If Rr≥a, it means that the combustion index Rr of the mixed gas in the current measurement cycle is greater than or equal to the combustion critical value a, that is, the combustion index is not within the normal value range, and the mixed gas meets the combustion standard;

[0090] If the instruction input by the user in the display layer is a combustion production instruction, the mixed gas in the current measurement cycle has reached the combustion standard, and the burner can be burned by ignition; if the instruction input by the user in the display layer is a non-combustion production instruction, it is necessary to control the mixed gas that has reached the combustion standard in the current measurement cycle, and the monitoring module issues a control command to the control layer;

[0091] Step S43: if Rr<a, it means that the combustion index Rr of the mixed gas in the current measurement cycle is less than the combustion critical value a, that is, the combustion index is within the normal value range, and the mixed gas is normal and does not meet the combustion standard;

[0092] If the instruction input by the user in the display layer is a combustion production instruction, the mixed gas in the current measurement cycle does not meet the combustion standard, and it is necessary to regulate the individual gas in the mixed gas, and the monitoring module issues a regulation command to the regulation layer; if the instruction input by the user in the display layer is a non-combustion production instruction, the mixed gas in the current measurement cycle needs to meet the user's production standard, and no additional regulation is required.

[0093] Wherein, a is the combustion critical value of the burner ignition mixed gas, and the monitoring equipment and the monitoring module both perform continuous periodic work.

[0094] The specific working process of the monitoring module and the judgment module in the processing layer includes the following steps:

[0095] Step S51: Each feed pipe of the burner is numbered, and the measurement cycle and cycle time are set. The flow measurement device stably obtains the flow value of the gas in each feed pipe, and the pressure measurement device stably obtains the pressure value of the gas in each feed pipe, and stably obtains the mixed gas pressure of the single gas in the mixed feed pipe. The user sets each parameter in the evaluation formula through the display layer;

[0096] Step S52: the monitoring module brings the values ​​transmitted by the pressure measuring device and the flow measuring device into the evaluation formula to obtain the combustion index of the burner after the single gas is mixed;

[0097] Step S53: The judgment module judges whether the combustion index of the mixed gas reaches the ignition combustion standard based on the production instructions entered by the user from the display layer, and outputs different control instructions to the control layer according to the judgment result. The control layer controls the flow input of a single gas according to the requirements.

[0098] In this embodiment, by carrying out a distributed design of the database, it is not only convenient for rapid response when reading data, but also realizes the elastic expansion of storage space, thereby improving the operating performance of the entire system; by establishing roles and permissions to manage and design user information, different roles are given to users to obtain different permissions, thereby realizing the rigor of the user in the system management process and the management efficiency of data information, and obtaining the combustion index of the mixed gas according to the evaluation formula to provide data support for the state determination of the mixed gas in the judgment module. The judgment module generates different application instructions according to the judgment results, thereby realizing the analysis and determination of the burner under different mixed gas composition states, and providing detailed and reliable data support for the regulation and management of the control layer.

[0099] Embodiment 2

[0100] Reference Figure 1 , Figure 3 , Figure 4 , Figure 5 and Figure 6 , Example 2 further illustrates a burner regulation and control system based on artificial intelligence proposed by the present invention.

[0101] A burner regulation and control system based on artificial intelligence, the air mixing regulation system includes a regulation layer; the regulation layer regulates the input speed of a single gas based on PID control of a neural network, and then feedback regulates the flow rate of a single gas according to the regulation of the input speed, so that the concentration of a single gas in the mixed gas reaches the relevant standard, and the regulation layer constructs a PID-speed regulation model, and the specific construction process of the PID-speed regulation model is as follows:

[0102] Step S61: Continuously discretize the cycle time s of a measurement cycle into multiple unit times t to construct a discrete control model.

[0103] (2)sv(s)=de(n)[v(t)+e(t)]

[0104] (3) v(t) = v(t-1) + Δv(t)

[0105] (4)Δv(t)=K B [e(t)-e(t-1)]+K J e(t)+K W [e(t)+e(t-2)-2e(t-1)] Among them, formula (2), formula (3) and formula (4) are all obtained by Discretize the deformation.

[0106] Formula (2) is used to show the expression of the gas input speed outputted in each iteration approaching the preset input speed;

[0107] Formula (3) is used to show the incremental control expression of iterative gas input speed;

[0108] Formula (4) is used to show the expression for obtaining Δv(t);

[0109] Wherein, sv(s) represents the expected transmission speed set when a single gas reaches the combustion concentration within the periodic measurement time s, v(t) represents the gas input speed output after each iteration, which is simplified as the iterative output speed, e(t) represents the absolute value of the deviation between the iterative output speed v(t) per unit time t and the expected transmission speed sv(s), v(t-1) represents the transmission speed value actually output at the previous moment of the current sampling moment t, Δv(t) represents the actual transmission speed difference between the current sampling moment and the previous sampling moment, e(t-1) represents the absolute value of the deviation within the previous sampling moment of unit time t, e(t-2) represents the absolute value of the deviation within the previous sampling moment of unit time t-1, de() represents the iterative function; n represents the number of iterations; K B is the proportionality coefficient; K J Indicates the integral coefficient, K W represents the differential coefficient.

[0110] Step S62: Based on the gas input speed output after each iteration, the proportionality coefficient K B , integral coefficient K J , differential coefficient K W Adaptive adjustment is performed, and finally the gas input speed actually output by the PID-speed control model is continuously approached to the expected transmission speed sv(s).

[0111] The control layer uses neural network to control the proportional coefficient K in PID control B , integral coefficient K J , differential coefficient KW Adaptive correction and adjustment are performed. The neural network consists of an input layer, a hidden layer, and an output layer. The proportional coefficient K is updated by updating the weights. B , integral coefficient K J , differential coefficient K W The specific process of adjusting and building the neural network model is as follows:

[0112] (5)

[0113] (6)

[0114] (7) Formula (5) is used to express the relationship between the input of the given input layer and the input and output of the hidden layer;

[0115] Among them, b represents the current iteration number, m() represents the activation function, It indicates the input of the given input layer with label j, o (1) represents the input of a given input layer, represents the input of the hidden layer labeled l, c (2) represents the input of the hidden layer, represents the output of the hidden layer labeled h, O (2) represents the output of the hidden layer, Represents the weight coefficient from the input layer to the hidden layer;

[0116] In addition, the number of input layers is set to M, and j is the number of any one of the M input layers, the number of input layers in the hidden layer is set to L, and l is the number of any one of the L hidden layers, the number of output layers in the hidden layer is set to H, and h is the number of any one of the H hidden layers, the number of output layers is set to 3, and k is the number of any one of the 3 output layers;

[0117] Formula (6) is used to show the input and output expressions of the output layer;

[0118] Among them, g() represents the activation function, represents the weight coefficient from the hidden layer to the output layer, Represents the output variable, corresponding to the three parameters controlled by PID. represents the input variable of the output layer, represents the input of the hidden layer labeled l, It indicates the input variable of the output layer when the label h takes the value of 1, 2, or 3;

[0119] Formula (7) is used for weighting coefficient and The update expression of , α is the inertia coefficient, η is the learning efficiency, is the intermediate variable, For the bth iteration and the b-1th iteration The weighted coefficient difference of For the bth iteration and the b-1th iteration The difference in weighting coefficients of and.

[0120] The control process of PID control based on neural network in the control layer includes the following steps:

[0121] Step S81: Determine the neural network structure, set the number of layers of the input layer, hidden layer and output layer respectively. The number of layers of the output layer in the neural network model constructed this time is 3, and set the proportional coefficient K of PID according to the PID-speed control model. B , integral coefficient K J , differential coefficient K W And other parameters to determine the initial value, select the appropriate inertia coefficient and learning efficiency;

[0122] Step S82: The flow measurement device collects the gas input speed output per unit time t, inputs the expected transmission speed value during the bobbin working process, calculates and obtains the absolute value of the deviation between the iterative output speed v(t-1) per unit time t-1 and the expected transmission speed sv(s), and calculates and obtains v(t) at the unit time t=1;

[0123] Step S83: Calculate the input and output of each node in each layer of the neural network through the neural network model to obtain the proportional coefficient K of the modified PID control B , integral coefficient K J , differential coefficient K W ;

[0124] Step S84: Calculate v(t+1) when t=t+1, and repeat steps S82 and S84.

[0125] Based on the above process, the PID-speed control model can be applied to the input speed control of any single gas in the mixed gas, or one or more of them can be controlled together. The speed before control is set to 1, and the expected transmission speed is set to 2, then v(t) = 1, sv(s) = 2, that is, e(t = 1. After one control, we can get

[0126] v(t+1)=v(t)+K B [e(t)-e(t-1)]+K J e(t)+K W [e(t)+e(t-2)-2e(t-1)]=1+K B ×1+K J ×1+KW ×1

[0127] , repeat in sequence until the expected transmission speed 2 is approached, where the proportional coefficient K B , integral coefficient K J , differential coefficient K W Regulated by the update of the neural network model.

[0128] In this embodiment, by constructing a PID-speed control model, the actual transmission speed and the expected transmission speed are jointly expressed by formulas, and by accessing a neural network model, the number of layers of each input layer, hidden layer and output layer is set based on the adaptive update of the weighting coefficients in the neural network model, thereby realizing automatic control of the input speed of a single gas by the PID-speed control model based on the neural network, improving the control performance of the actual transmission speed approaching the expected transmission speed, and improving the speed, accuracy and stability in the speed approximation process.

[0129] Embodiment 3

[0130] Reference Figure 1 and Figure 6 , Example 3 further illustrates the burner regulation and control system based on artificial intelligence proposed by the present invention.

[0131] A burner regulation and control system based on artificial intelligence, the air mixing regulation system includes a data layer, a processing layer, a regulation layer and a display layer; the technical solution of the air mixing regulation system mainly includes the following steps:

[0132] Step S1: The user performs identity verification through the client's Web browsing page to determine whether to allow login to the system, and obtains corresponding access and management permissions according to the role assigned by the user identity. The user manages the gas parameters and control parameters of the burner by carrying the role;

[0133] Step S2: The burner is continuously monitored through the pressure measuring device and the flow measuring device. The monitoring module of the processing layer obtains the combustion index of the mixed gas through the evaluation formula. The judgment module generates different application instructions according to the judgment result. The control layer performs relevant processing according to the signal information sent by the application instruction.

[0134] Step S3: The control layer establishes an expression for the gas input speed through the PID-speed control model, and connects the PID-speed control model to the neural network model to adjust the proportional coefficient K of the PID control. B , integral coefficient K J , differential coefficient K W Make adaptive adjustments;

[0135] Step S4: Repeat steps S2 to S4.

[0136] The air mixing and conditioning system is applied to the cloud processing platform, the data layer is applied to the cloud, the central processing unit is set on the cloud processing platform and adopts an integrated information processing system. The processing layer and the control layer perform cloud processing based on the central processing unit. The central processing unit is connected to the monitoring equipment, the burner and the client through a wireless network or wired transmission. The display layer displays information through the Web page program of the output device, outputs the processing results of the central processing unit, and provides information for management personnel to receive. The output layer adopts a multi-client information architecture. Users access the system through the client and have permission to access information according to the assigned role.

[0137] According to another aspect of the present application, a monitoring device is provided. The monitoring device may include one or more processors and one or more memories. The memories may store computer readable codes, which, when executed by the one or more processors, may perform intelligent regulation of air mixing.

[0138] The central processing unit may also refer to a computer, and a computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by the processor, intelligent regulation of air mixing may be performed. The storage medium includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0139] In addition, according to the embodiments of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, a real-time control method based on supine body measurement. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0140] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or a wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0141] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0142] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0143] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0144] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0145] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0146] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A burner regulation control system based on artificial intelligence, characterized in that: The air mixing and conditioning system comprises a data layer, a processing layer, a control layer and a display layer; The data layer includes a database and a call group. The data layer and the processing layer are designed based on the NET open source architecture. The database includes a cache of information related to the processing layer and the device information of the burner, as well as user management information, and performs distributed storage and call management. The processing layer includes a monitoring module and a judgment module. The monitoring module obtains the combustion index of the mixed gas through an evaluation formula to provide data support for the judgment of the judgment module. The judgment module generates different control instructions according to the judgment results. The control layer includes a PID-speed control model and a neural network model. The proportional coefficient K of the PID-speed control model is adjusted by the neural network model. B , integral coefficient K J , differential coefficient K W Adaptively adjust the input speed of a single gas in the mixed gas to the desired transmission speed; The display layer is used for input and output of application instructions and for terminal page display. The application instructions of the display layer include production instructions.

2. The artificial intelligence-based burner regulation and control system according to claim 1, characterized in that: The user management information includes user identity authentication and authority management. User identity authentication is applied to user login. The calling group verifies the identity information entered by the user when logging in with the identity information stored in the database to determine whether the user has the login authority. Permission management designs roles through a permission model, assigns corresponding roles based on the user's authenticated identity, and obtains corresponding permissions based on the role.

3. The artificial intelligence-based burner adjustment and control system according to claim 1, characterized in that: The monitoring module is connected to the monitoring device. The monitoring module processes the flow values ​​on each feed pipeline obtained by the monitoring device. The monitoring device includes a flow measurement device and a pressure measurement device. The fuel state of the burner at the current moment is obtained according to the numerical processing of the monitoring module. The specific working process of the monitoring module includes the following steps: Step S31: The pressure measuring device transmits the pressure of the gas in each feed pipeline to the monitoring module through the wireless network, and the flow measuring device transmits the flow of the gas in each feed pipeline to the monitoring module through the wireless network, obtains the flow of a single gas in the feed pipeline and the flow of the mixed gas, and obtains the combustion index of the mixed gas according to the evaluation formula: Formula (1) Formula (1) is used to obtain the combustion index Rr of the burner after mixing a single gas; Where μ represents the turbulence intensity coefficient in the mixed gas, P i represents the pressure of a single gas with label i, Pe represents the overall pressure of the mixed gas after all the single gases are mixed, m i represents the mass of a single gas with label i, s represents the cycle time set for a measurement cycle, and q i Indicates the flow rate of a single gas with label i, λ i represents the concentration ratio of the single gas with label i, T represents the average temperature of the mixed gas after all the single gases are mixed, ε represents the ratio of the average temperature, i represents the label of any single gas in the mixed gas, and I represents the number of all the single gases in the mixed gas; Step S32: the monitoring module transmits the combustion index Rr of the mixed single gases to the judgment module.

4. The artificial intelligence-based burner adjustment and control system according to claim 1, characterized in that: The judgment module is used to judge the combustion degree of the mixed gas, output different control instructions according to the judgment result, and perform feedback control on the production instructions issued in the display layer. The specific working process includes the following steps: Step S41: the judgment module receives the combustion index Rr of the mixed single gases in the monitoring module; Step S42: If Rr≥a, it means that the combustion index Rr of the mixed gas in the current measurement cycle is greater than or equal to the combustion critical value a, that is, the combustion index is not within the normal value range, and the mixed gas meets the combustion standard; If the instruction input by the user in the display layer is a combustion production instruction, the mixed gas in the current measurement cycle has reached the combustion standard, and the burner can be burned by ignition; if the instruction input by the user in the display layer is a non-combustion production instruction, it is necessary to control the mixed gas that has reached the combustion standard in the current measurement cycle, and the monitoring module issues a control command to the control layer; Step S43: if Rr<a, it means that the combustion index Rr of the mixed gas in the current measurement cycle is less than the combustion critical value a, that is, the combustion index is within the normal value range, and the mixed gas is normal and does not meet the combustion standard; If the instruction input by the user in the display layer is a combustion production instruction, the mixed gas in the current measurement cycle does not meet the combustion standard, and it is necessary to regulate the individual gas in the mixed gas, and the monitoring module issues a regulation command to the regulation layer; if the instruction input by the user in the display layer is a non-combustion production instruction, the mixed gas in the current measurement cycle needs to meet the user's production standard, and no additional regulation is required.

5. The artificial intelligence-based burner adjustment and control system according to claim 1, characterized in that: The specific working process of the monitoring module and the judgment module in the processing layer includes the following steps: Step S51: Each feed pipe of the burner is numbered, and the measurement cycle and cycle time are set. The flow measurement device stably obtains the flow value of the gas in each feed pipe, and the pressure measurement device stably obtains the pressure value of the gas in each feed pipe, and stably obtains the mixed gas pressure of the single gas in the mixed feed pipe. The user sets each parameter in the evaluation formula through the display layer; Step S52: the monitoring module brings the values ​​transmitted by the pressure measuring device and the flow measuring device into the evaluation formula to obtain the combustion index of the burner after the single gas is mixed; Step S53: The judgment module judges whether the combustion index of the mixed gas reaches the ignition combustion standard based on the production instructions entered by the user from the display layer, and outputs different control instructions to the control layer according to the judgment result. The control layer controls the flow input of a single gas according to the requirements.

6. The artificial intelligence-based burner adjustment and control system according to claim 5, characterized in that: The control layer controls the input speed of a single gas based on the PID control of the neural network, and then adjusts the flow rate of the single gas by feedback according to the control of the input speed, so that the concentration of the single gas in the mixed gas reaches the relevant standard. The control layer constructs a PID-speed control model. The specific construction process of the PID-speed control model is as follows: Step S61: Continuously discretize the cycle time s of a measurement cycle into multiple unit times t to construct a discrete control model. (2)sv(s)=de(n)[v(t)+e(t)] (3) v(t) = v(t-1) + Δv(t) (4)Δv(t)=K B [e(t)-e(t-1)]+K J e(t)+K W [e(t)+e(t-2)-2e(t-1)] Formula (2) is used to show the expression of the gas input speed outputted in each iteration approaching the preset input speed; Formula (3) is used to show the incremental control expression of the iterative gas input speed; Formula (4) is used to show the expression for obtaining Δv(t); Wherein, sv(s) represents the expected transmission speed set when a single gas reaches the combustion concentration within the periodic measurement time s, v(t) represents the gas input speed output after each iteration, which is simplified as the iterative output speed, e(t) represents the absolute value of the deviation between the iterative output speed v(t) per unit time t and the expected transmission speed sv(s), v(t-1) represents the transmission speed value actually output at the previous moment of the current sampling moment t, Δv(t) represents the actual transmission speed difference between the current sampling moment and the previous sampling moment, e(t-1) represents the absolute value of the deviation within the previous sampling moment of unit time t, e(t-2) represents the absolute value of the deviation within the previous sampling moment of unit time t-1, de() represents the iterative function; n represents the number of iterations; K B is the proportionality coefficient; K J Indicates the integral coefficient, K W represents the differential coefficient; Step S62: Based on the gas input speed output after each iteration, the proportionality coefficient K B , integral coefficient K J , differential coefficient K W Adaptive adjustment is performed, and finally the gas input speed actually output by the PID-speed control model is continuously approached to the expected transmission speed sv(s).

7. The artificial intelligence-based burner adjustment and control system according to claim 6, characterized in that: The control layer is based on the proportional coefficient K in the PID control of the neural network B , integral coefficient K J , differential coefficient K W Adaptive correction and adjustment are performed. The neural network consists of an input layer, a hidden layer, and an output layer. The proportional coefficient K is updated by updating the weights. B , integral coefficient K J , differential coefficient K W The specific process of adjusting and building the neural network model is as follows: (5) (6) (7) Formula (5) is used to express the relationship between the input of the given input layer and the input and output of the hidden layer; Among them, b represents the current iteration number, m() represents the activation function, It indicates the input of the given input layer with label j, o (1) represents the input of a given input layer, represents the input of the hidden layer labeled l, c (2) represents the input of the hidden layer, represents the output of the hidden layer labeled h, O (2) represents the output of the hidden layer, Represents the weight coefficient from the input layer to the hidden layer; In addition, the number of input layers is set to M, and j is the number of any one of the M input layers, the number of input layers in the hidden layer is set to L, and l is the number of any one of the L hidden layers, the number of output layers in the hidden layer is set to H, and h is the number of any one of the H hidden layers, the number of output layers is set to 3, and k is the number of any one of the 3 output layers; Formula (6) is used to show the input and output expressions of the output layer; Among them, g() represents the activation function, represents the weight coefficient from the hidden layer to the output layer, Represents the output variable, corresponding to the three parameters of PID control, represents the input variable of the output layer, represents the input of the hidden layer labeled l, It indicates the input variable of the output layer when the label h takes the value of 1, 2, or 3; Formula (7) is used for weighting coefficient and The update expression of , α is the inertia coefficient, η is the learning efficiency, is the intermediate variable, For the bth iteration and the b-1th iteration The weighted coefficient difference of For the bth iteration and the b-1th iteration The difference in weighting coefficients of and.

8. The artificial intelligence-based burner adjustment and control system according to claim 7, characterized in that: The control process of the control layer for PID control based on the neural network includes the following steps: Step S81: Determine the neural network structure, set the number of layers of the input layer, hidden layer and output layer respectively. The number of layers of the output layer in the neural network model constructed this time is 3, and set the proportional coefficient K of PID according to the PID-speed control model. B , integral coefficient K J , differential coefficient K W And other parameters to determine the initial value, select the appropriate inertia coefficient and learning efficiency; Step S82: The flow measurement device collects the gas input speed output per unit time t, inputs the expected transmission speed value during the bobbin working process, calculates and obtains the absolute value of the deviation between the iterative output speed v(t-1) per unit time t-1 and the expected transmission speed sv(s), and calculates and obtains v(t) at the unit time t=1; Step S83: Calculate the input and output of each node in each layer of the neural network through the neural network model to obtain the proportional coefficient K of the corrected PID control B , integral coefficient K J , differential coefficient K W ; Step S84: Calculate v(t+1) when t=t+1, and repeat steps S82 and S84.

9. The artificial intelligence-based burner adjustment and control system according to claim 8, characterized in that: The technical solution of the air mixing and conditioning system mainly includes the following steps: Step S1: The user performs identity verification through the client's Web browsing page to determine whether to allow login to the system, and obtains corresponding access and management permissions according to the role assigned by the user identity. The user manages the gas parameters and control parameters of the burner by carrying the role; Step S2: The burner is continuously monitored through the pressure measuring device and the flow measuring device. The monitoring module of the processing layer obtains the combustion index of the mixed gas through the evaluation formula. The judgment module generates different application instructions according to the judgment result. The control layer performs relevant processing according to the signal information sent by the application instruction. Step S3: The control layer establishes an expression for the gas input speed through the PID-speed control model, and connects the PID-speed control model to the neural network model to adjust the proportional coefficient K of the PID control. B , integral coefficient K J , differential coefficient K W Make adaptive adjustments; Step S4: Repeat steps S2 to S4.

10. The artificial intelligence-based burner adjustment and control system according to claim 9, characterized in that: The air mixing and conditioning system is applied to a cloud processing platform, the data layer is applied to the cloud, the central processing unit is arranged on the cloud processing platform and adopts an integrated information processing system, the processing layer and the control layer perform cloud processing based on the central processing unit, the central processing unit is connected to the monitoring equipment, the burner and the client through a wireless network or wired transmission, the display layer displays information through the Web page program of the output device, outputs the processing results of the central processing unit, and provides information for management personnel to receive, the output layer adopts a multi-client information architecture, the user accesses the system through the client, and has permission to access the information according to the assigned role.

Images