Torch combustion control method, device and equipment based on meteorological parameters and medium

Abstract

The invention relates to the technical field of torch combustion. The invention discloses a torch combustion control method, device and equipment based on meteorological parameters and a medium. The method comprises the following steps: acquiring meteorological parameters including at least one of air temperature, air pressure, wind direction, wind power, illumination and rainfall; the meteorological parameters serve as input values to be input into a pre-trained neural network model, and the output value of the pre-trained neural network model is the estimated opening degree of each adjusting valve; and adjusting the opening degree of the corresponding adjusting valve according to the estimated opening degree of each adjusting valve. The meteorological parameters are input into the pre-trained neural network model to obtain the opening estimated value of the adjusting valve, the adjusting valve in the combustion system is adjusted according to the opening estimated value of the adjusting valve so as to control the flame intensity, automatic adjustment of the torch combustion system is achieved, time and labor are saved, efficiency is high, and the effect is good.

Description

Technical Field

[0001] The present application relates to the technical field of flare combustion, and in particular to a flare combustion control method, device, equipment and medium based on meteorological parameters. Background Art

[0002] At the opening ceremony of the Games, the main torch is the focus of everyone's attention. The existing torch combustion is generally achieved by manually adjusting the angle of the gas supply ball valve or controlling the opening of the electric regulating valve, opening or closing the gas supply branch ball valve or solenoid valve to increase or decrease the flame intensity. Due to the long duration of the Games and the changes in the weather, the adjustment of the torch combustion system is time-consuming and laborious, with low efficiency and poor effect. Summary of the invention

[0003] In view of the above situation, the embodiments of the present application provide a flare combustion control method, device, equipment and medium based on meteorological parameters, aiming to solve the above problems or at least partially solve the above problems.

[0004] In a first aspect, a flare combustion control method based on meteorological parameters is provided, comprising:

[0005] Obtaining meteorological parameters, including at least one of temperature, air pressure, wind direction, wind speed, light, and rainfall;

[0006] Inputting the meteorological parameters as input values ​​into a pre-trained neural network model, wherein the output value of the pre-trained neural network model is the estimated opening of each regulating valve;

[0007] The opening of the corresponding regulating valve is adjusted according to the estimated opening of each regulating valve.

[0008] Preferably, adjusting the corresponding regulating valve according to the estimated opening of each regulating valve includes:

[0009] According to the estimated opening of the regulating valve and based on the mapping relationship between the opening of the regulating valve and the fuel pipeline pressure, a pipeline pressure setting value of the fuel pipeline where the regulating valve is located is obtained;

[0010] The opening of the regulating valve is adjusted according to the pipeline pressure setting value of the fuel pipeline where the regulating valve is located.

[0011] Preferably, the method further comprises:

[0012] When training the neural network model, the weights and thresholds of the neural network model are updated using the flood optimization algorithm, and the updating process includes:

[0013] Get the parameters of the flood optimization algorithm;

[0014] Creating an initial population of a flood optimization algorithm according to the meteorological parameters, initial weight values, and initial threshold values;

[0015] Determine the initial position for the start of iteration, and calculate the fitness of individuals in the initial population in each iteration, where the fitness is the performance of the neural network model under the current weights and thresholds;

[0016] According to the fitness of individuals in the initial population, the initial population is updated by a flood optimization algorithm mechanism to obtain an updated population, wherein the individuals in the updated population are located at new positions;

[0017] Based on the pre-set judgment rules, the individual's position is updated;

[0018] Based on the preset output conditions, the weight and threshold represented by the global optimal position are output.

[0019] Preferably, according to the fitness of individuals in the initial population, the initial population is updated by a flood optimization algorithm mechanism to obtain an updated population, wherein the individuals in the updated population are located at new positions, including:

[0020] Calculating individual optimization costs according to the fitness of individuals in the initial population;

[0021] Determining an update type of the individual position according to the individual optimization cost based on a random number of a current iteration and a random number of a previous iteration;

[0022] Based on the determined update type of the individual position, the new position of the individual is calculated to obtain the updated population.

[0023] Preferably, based on a preset judgment rule, the location of the individual is updated, including:

[0024] Based on the preset optimization function, the optimization function value when the individual is at the new position and the optimization function value when the individual is at the old position are calculated respectively;

[0025] When the optimization function value when the individual is at the new position is less than the optimization function value when the individual is at the old position, it is determined that the position of the individual is replaced from the old position to the new position, and the optimization function value when the individual is at the best position is replaced by the optimization function value when the individual is at the new position;

[0026] When the optimization function value when the individual is at the old position is less than the optimization function value at the optimal position, it is determined that the position of the individual is replaced from the optimal position to the old position, and the optimization function value of the individual at the optimal position is replaced by the optimization function value when the individual is at the old position.

[0027] Preferably, based on a preset judgment rule, updating the location of the individual also includes:

[0028] Calculate the probability of water volume increasing or decreasing based on the random number of the current iteration;

[0029] According to the random number of the current iteration and the probability comparison result of water volume increase or decrease, the optimization function value when the individual is located at the new position is calculated;

[0030] When the optimization function value when the individual is at the new position is less than the optimization function value at the optimal position, it is determined that the position of the individual is replaced from the optimal position to the new position, and the optimization function value of the individual at the optimal position is replaced by the optimization function value when the individual is at the new position.

[0031] Preferably, the method further comprises:

[0032] Obtain fuel concentration detection value;

[0033] Based on the preset fuel concentration preset value, the start or stop of the explosion-proof fan is controlled according to the fuel concentration detection value.

[0034] In a second aspect, a flare combustion control device based on meteorological parameters is provided, comprising:

[0035] An acquisition module, used to acquire meteorological parameters, including at least one of temperature, air pressure, wind direction, wind force, light, and rainfall;

[0036] A calculation module, used for inputting the meteorological parameters as input values ​​into a pre-trained neural network model, wherein the output value of the pre-trained neural network model is an estimated opening degree of each regulating valve;

[0037] The adjustment module is used to adjust the opening of the corresponding regulating valve according to the estimated opening of each regulating valve.

[0038] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the flare combustion control method based on meteorological parameters as described in the first aspect are implemented.

[0039] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the steps of the flare combustion control method based on meteorological parameters as described in the first aspect are implemented.

[0040] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0041] The present application obtains an estimated value of the opening of a regulating valve by inputting meteorological parameters into a pre-trained neural network model, and adjusts the regulating valve in the combustion system according to the estimated value of the opening of the regulating valve to control the intensity of the flame, thereby realizing automated adjustment of the flare combustion system, saving time and effort, and having high efficiency and good effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0043] Figure 1 This is a schematic diagram of the principle of a flare combustion control system based on meteorological parameters in one embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the principle of a PLC control cabinet in one embodiment of the present invention;

[0045] Figure 3 This is a schematic diagram of the data transmission principle between a PLC control cabinet and an industrial computer in one embodiment of the present invention;

[0046] Figure 4 It is a schematic diagram of an application environment of a flare combustion control method based on meteorological parameters in one embodiment of the present invention;

[0047] Figure 5 It is a flow chart of a flare combustion control method based on meteorological parameters in one embodiment of the present invention;

[0048] Figure 6 It is a schematic diagram of a process of updating weights and thresholds in a neural network model in one embodiment of the present invention;

[0049] Figure 7 is another schematic diagram of a flow chart of updating weights and thresholds in a neural network model in one embodiment of the present invention;

[0050] Figure 8 is another schematic diagram of a flow chart of updating weights and thresholds in a neural network model in one embodiment of the present invention;

[0051] Fig. 9 It is a structural schematic diagram of a flare combustion control device based on meteorological parameters in one embodiment of the present invention;

[0052] Fig.10 is a schematic diagram of a structure of a computer device in one embodiment of the present invention;

[0053] Fig.11 It is another structural schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0055] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such use is interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the term "including" and its variants are to be interpreted as open-ended terms meaning "including but not limited to".

[0056] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0057] Embodiment 1

[0058] like Figure 1 and Figure 2 As shown, this embodiment introduces a flare combustion system based on meteorological parameters, including a control room 100, a gas equipment room 200, a combustion device 300, a meteorological station 400, and a nitrogen room 500. The control room 100 is provided with a PLC control cabinet 110 and a MCGS configuration software monitoring system 120, and the PLC control cabinet 110 and the MCGS configuration software monitoring system 120 are electrically connected. Figure 2 and Figure 3 As shown, the PLC control cabinet 110 includes a first module 111 , a second module 112 , a third module 113 , and a fourth module 114 .

[0059] A gas equipment room 200 is provided with a fuel supply device, an explosion-proof fan 250 communicatively connected to a PLC control cabinet 110, an explosion-proof junction box 230 electrically connected to the PLC control cabinet 110, a fuel concentration detection device 240, and a fuel pipeline connected to the fuel supply device. The fuel pipeline includes a first outer ring gas pipeline connected to an outer ring burner and a second outer ring gas pipeline, a middle ring gas pipeline connected to a middle ring burner, an inner ring gas pipeline connected to an inner ring burner, an inner ring long open flame gas pipeline connected to an inner ring long open flame burner, and a middle ring long open flame gas pipeline connected to a middle ring long open flame burner. The first outer ring gas pipeline is provided with a first solenoid valve 221 and a first electric regulating valve 211, the second outer ring gas pipeline is provided with a second solenoid valve 222 and a second electric regulating valve 212, the middle ring gas pipeline is provided with a third solenoid valve 223 and a third electric regulating valve 213, the inner ring gas pipeline is provided with a fourth solenoid valve 224 and a fourth electric regulating valve 214, the inner ring long open flame gas pipeline is provided with a fifth solenoid valve 225, and the middle ring long open flame gas pipeline is provided with a sixth solenoid valve 226. The first solenoid valve 221, the second solenoid valve 222, the third solenoid valve 223, the fourth solenoid valve 224, the fifth solenoid valve 225, the sixth solenoid valve 226, the first electric regulating valve 211, the second electric regulating valve 212, the third electric regulating valve 213 and the fourth electric regulating valve 214 are electrically connected to the PLC control cabinet 110.

[0060] The combustion device 300 includes burners connected to the fuel pipeline, an ignition equipment box 310 electrically connected to the PLC control cabinet 110, a fire detection equipment box 320, a temperature transmitter box 330, a combustible gas detector 340, and a video equipment box 350. The ignition equipment box 310 is electrically connected to the igniter, the fire detection equipment box 320 is electrically connected to the flame detection device, the temperature transmitter box 330 is electrically connected to the temperature detection element, and the video equipment box 350 is electrically connected to the camera 351. To obtain flame detection information and temperature detection information about the combustion device. The camera 351 is set at the lower end of the burner, and the burner equipment status and the combustion flame status can be viewed in real time.

[0061] The weather station 400 includes an anemometer 410, which is arranged near the burner to obtain the wind force and direction information of the flame; it also includes an air temperature and pressure sensor 420, an illumination sensor 430, a rainfall sensor 440, and a meteorological equipment box 450. The air temperature and pressure sensor 420, the illumination sensor 430, the rainfall sensor 440, and the wind speed and direction sensor 410 are electrically connected to the meteorological equipment box 450, respectively, and the meteorological equipment box 450 is electrically connected to the PLC control cabinet 110. The meteorological data collected by the weather station 400 can also be transmitted to the industrial computer 120 through the Internet of Things module. The anemometer 410 is made of aluminum alloy, waterproof and shockproof, small in size, simple to install, high in detection accuracy, low in power consumption, and the wind direction is defined according to the 16-azimuth map. The illumination sensor 430 uses a photosensitive detector with high sensitivity, cooperates with a high-precision linear amplifier circuit, and has a delicate structure. The core component of the rain sensor 440, the tipping bucket, adopts a three-dimensional streamlined shape, which makes the tipping bucket transfer water more smoothly and easy to clean.

[0062] The nitrogen room 500 is provided with a nitrogen bottle group 520. The nitrogen source device 520 is connected to the fuel pipeline through a nitrogen valve 510. The nitrogen room 500 is skid-mounted. In this embodiment, nitrogen purge and replacement are performed before the fuel pipeline is ventilated to exhaust the air in the fuel pipeline to prevent the fuel pipeline from being mixed with air and deflagrated during ignition. The fuel in the fuel pipeline should also be purged after the torch is extinguished to prevent accidents caused by the remaining fuel in the fuel pipeline.

[0063] Specifically, the flare combustion system control parameters can be set through the MCGS configuration software monitoring system 120 parameter setting interface. The control parameters include igniter ignition time setting, gas pipeline temperature and pressure upper limit setting, meteorological parameter upper limit setting, nitrogen purge time setting, fuel concentration upper limit setting, system automatic / manual selection, and matrix combustion nozzle solenoid valve point enable.

[0064] Specifically, through the MCGS configuration software monitoring system 120 system monitoring interface, manual testing and start and stop of each igniter, solenoid valve, electric control valve, and nitrogen valve can be performed.

[0065] Specifically, nitrogen can be used to purge the pipeline of the flare combustion system through the system monitoring interface of the MCGS configuration software monitoring system 120, and the PLC control cabinet 110 completes the opening and closing of the nitrogen valve according to the set time.

[0066] Specifically, the PLC control cabinet 110 can automatically start and stop the explosion-proof fan 250 between the valve groups according to the detection signal of the gas detector 240 and the upper limit setting of the fuel concentration.

[0067] Specifically, the ignition equipment box 310 can be controlled to ignite the long-open flame through the system monitoring interface of the MCGS configuration software monitoring system 120 before the torch is ignited, and the flame detection can be performed through the flame detection sensor. After the long-open flame is ignited and the flame detection has a signal, the ignition of the torch combustion system is completed by the system monitoring interface of the MCGS configuration software monitoring system 120. During the ignition, the high fire / low fire / extinguishing can be switched by the buttons on the system monitoring interface of the MCGS configuration software monitoring system 120. The switching work is completed by the PLC control cabinet 110 by adjusting the opening of the first electric regulating valve 211, the second electric regulating valve 212, and the third electric regulating valve 213 set on the outer ring gas pipeline and the middle ring gas pipeline, and closing the first solenoid valve 221, the second solenoid valve 222, and the third solenoid valve 223 corresponding to the high fire.

[0068] Specifically, the MCGS configuration software monitoring system 120 system monitoring interface process animation displays the working status of each device in the flare combustion system and the readings of each sensor, which is clear and intuitive, allowing operators or operation support personnel to more conveniently monitor the operating status of the flare combustion system.

[0069] Specifically, during the combustion process of the flare combustion system, based on the fuel pipeline pressure setting value, the PLC control cabinet 110 controls the opening of the first electric regulating valve 211, the second electric regulating valve 212, the third electric regulating valve 213 and the fourth electric regulating valve 214 in real time through PID control, so that the flare combustion posture and flame height can achieve a viewing effect.

[0070] Specifically, during the low-fire operation period during the competition period, the combustion mode is the low-fire mode. At this time, the gas flow rate is small, the first solenoid valve 221, the second solenoid valve 222, and the third solenoid valve 223 are all closed, and the fourth solenoid valve 224, the fifth solenoid valve 225, the sixth solenoid valve 226, the first electric regulating valve 211, the second electric regulating valve 212, the third electric regulating valve 213 and the fourth electric regulating valve 214 are working to ensure normal combustion of the low fire and achieve the purpose of saving gas.

[0071] Specifically, through the system alarm interface of the MCGS configuration software monitoring system 120, the fault alarm status of each equipment in the flare combustion system can be recorded in real time, including the opening of the first electric regulating valve 211, the second electric regulating valve 212, the third electric regulating valve 213 and the fourth electric regulating valve 214, fire detection failure, explosion-proof fan 250 failure, and over-limit readings of various sensors.

[0072] Specifically, during the combustion of the flare combustion system, the MCGS configuration software monitoring system 120 saves the data of various operating parameters and sensor readings in real time. The working status and trend analysis of the gas flare combustion system can be understood through data reports, real-time curves and historical curves, and daily reports can be viewed on key data, and data can be exported for processing and analysis.

[0073] Specifically, the communication method in the flare combustion control system uses a local area network, and the control signal does not rely on an external network or GPRS, which can effectively prevent network attacks;

[0074] Specifically, the torch combustion control system can perform matrix control of the first solenoid valve 221, the second solenoid valve 222, the third solenoid valve 223, and the fourth solenoid valve 224 according to the wind direction and wind speed. That is, the solenoid valves at the windward locations are closed accordingly to reduce the amount of flame drift.

[0075] Specifically, the torch combustion control system can adjust the first electric regulating valve 211, the second electric regulating valve 212, the third electric regulating valve 213 and the fourth electric regulating valve 214 according to the light intensity, so as to achieve both flame viewing and energy saving.

[0076] In this embodiment, meteorological parameters are introduced into the flare combustion control system, and the electric regulating valve in the system is adjusted according to future weather forecast information and real-time collected meteorological parameters, especially pre-processing of countermeasures for extreme weather, and enhancing the flare combustion system's ability to resist strong winds and heavy rains.

[0077] Embodiment 2

[0078] The flare combustion control method based on meteorological parameters provided by the embodiment of the present invention can be applied in the following aspects: Figure 4 In the application environment, the device communicates with the server through the network. The server can obtain meteorological parameters through the device, including at least one of temperature, air pressure, wind direction, wind force, light, and rainfall; the meteorological parameters are input into the pre-trained neural network model as input values, and the output value of the pre-trained neural network model is the estimated opening of each regulating valve; wherein, when training the neural network model, the weight and threshold of the neural network model are updated using the flood optimization algorithm; and the opening of the corresponding regulating valve is adjusted according to the estimated opening of each regulating valve. The present application obtains the estimated value of the opening of the regulating valve by inputting the meteorological parameters into the pre-trained neural network model, and adjusts the regulating valve in the combustion system according to the estimated value of the opening of the regulating valve to control the intensity of the flame, thereby realizing the automatic adjustment of the torch combustion system, saving time and labor, and having high efficiency and good effect. wherein, the device can be but is not limited to various sensors. The server can be implemented by an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0079] See also Figure 5 As shown, Figure 5A schematic flow chart of a flare combustion control method based on meteorological parameters provided in an embodiment of the present invention comprises the following steps:

[0080] S10: Obtain meteorological parameters, including at least one of temperature, air pressure, wind direction, wind speed, light, and rainfall.

[0081] Specifically, wind direction and wind speed are obtained by an anemometer 410 , air temperature and air pressure are obtained by an air temperature and air pressure sensor 420 , light is obtained by a light sensor 430 , and rainfall is obtained by a rainfall sensor 440 .

[0082] S20: Inputting the meteorological parameters as input values ​​into a pre-trained neural network model, the output value of the pre-trained neural network model being the estimated opening of each regulating valve.

[0083] Specifically, the neural network model in the present application adopts a three-layer BP neural network model, and the historical combustion data of the flare combustion system is input into the BP neural network model for training to obtain a pre-trained neural network model.

[0084] The input of the BP neural network model during the training process is: historical meteorological parameter data, and the output is the estimated opening of each regulating valve. The regulating valves include a first electric regulating valve 211 , a second electric regulating valve 212 , a third electric regulating valve 213 and a fourth electric regulating valve 214 .

[0085] During the training process, error analysis is performed each time based on the training results and the historical control valve opening data, and then the weights and thresholds are adjusted to gradually make the training results consistent with the historical control valve opening data.

[0086] During training, according to the selected BP neural network structure, first give the initial value of each layer weight and threshold, select the learning rate and inertia coefficient, and then calculate the opening of each regulating valve in the output layer. Figure 8 .

[0087] When applied, by inputting meteorological parameters into the trained neural network model, the opening of each regulating valve can be output, and each regulating valve can be adjusted in real time according to the opening of each regulating valve, thereby achieving the optimal control purpose of the system.

[0088] S30: adjusting the opening of the corresponding regulating valve according to the estimated opening of each regulating valve.

[0089] Specifically, they include:

[0090] S31: obtaining a pipeline pressure setting value of the fuel pipeline where the regulating valve is located according to the estimated opening of the regulating valve and a mapping relationship between the opening of the regulating valve and the fuel pipeline pressure;

[0091] S32: adjusting the opening of the regulating valve according to a pipeline pressure setting value of the fuel pipeline where the regulating valve is located.

[0092] Specifically, in this embodiment, the neural network model is set in MATLAB, and the DDE communication protocol is used to realize data exchange between the MCGS configuration software monitoring system 120 and MATLAB. During the combustion process of the flare combustion system, MATLAB performs BP neural network calculations based on meteorological parameters to obtain the estimated opening of each regulating valve. The PID control system obtains the pipeline pressure setting value of the fuel pipeline where the regulating valve is located based on the estimated opening of the regulating valve and the mapping relationship between the regulating valve opening and the fuel pipeline pressure, and sends it to the PLC control cabinet 110 by the MCGS configuration software monitoring system 120. The PLC control cabinet 110 performs PID control, and adjusts the opening of each electric regulating valve according to the pipeline pressure setting value of the fuel pipeline where the regulating valve is located, so that the combustion posture and flame height of the flare combustion system are more ornamental.

[0093] In this implementation, the six system parameters of meteorological parameters, namely wind direction, wind force, temperature, air pressure, light and rainfall, are used as the input vector X=(x1, x2, x3..., x6) of the BP neural network, and the output vector is the regulating valve adjustment opening gear Y=(y1, y2, y3..., y9), so as to establish a mapping relationship between the system parameters and the fuel flow regulation. The meteorological parameters and the corresponding weather conditions are used as learning samples. The sample data is usually divided into two parts, one part participates in the neural network training, and the other part participates in the testing and verification of the neural network. The BP neural network calculation is completed through MATLAB programming to obtain the better PID control parameters required by the PLC control cabinet 110. In this implementation, the regulating valve adjustment opening gear sets 9 gears of 0-100% opening, realizing the fine adjustment of the regulating valve.

[0094] Furthermore, the weights and thresholds of the neural network model are updated by using a flood optimization algorithm when training the neural network model. The flood optimization algorithm is selected to optimize the weights and thresholds of the BP neural network, thereby improving the generalization ability and learning performance of the BP neural network, thereby improving the overall search efficiency of the BP neural network and obtaining better PID adjustment parameters. The optimization algorithm is inspired by the complex movement and flow pattern of water masses. It has a unique algorithm structure and a novel iterative method, and has strong adaptive optimization capabilities.

[0095] Further, the flood optimization algorithm is introduced as follows:

[0096] In the normal moving phase, the optimization algorithm revolves around S j -S i And around S best Adjust the formula to:

[0097]

[0098] Among them, rand generates a random number in the range [0,1], j is the dimension of the problem in D-dimensional space, S j is the jth random individual in the population.

[0099] As the volume of water in the river increases, flooding and turbulence may occur, so the water consumption coefficient or water flow is modeled as follows:

[0100]

[0101] Where Pk is the water consumption coefficient, Iter max is the maximum number of iterations, and Iter is the current number of iterations;

[0102] Flood events occur randomly, with a probability of a random value. During the flood, the rand value is the water mass movement equation shown in equation (3). It can be seen that the water mass randomly flows from the pond to the other side of the area:

[0103]

[0104] In the formula, randn is a normal distribution, whose value varies from negative infinity to positive infinity and is one-dimensional. max and S min for the upper and lower bounds;

[0105] The permeability of the soil will affect the infiltration and diffusion of water, thereby reducing the possibility of flooding. The cost equation of the optimization function is used to model the cost of different group members. The lower the cost of the objective function, the higher the probability of flooding. For the i-th individual in the population, this individual optimization cost is expressed as:

[0106]

[0107] In the formula, f max and f min The best and worst values ​​of the optimization function found for the current iteration.

[0108] In the optimization algorithm, the probability of flooding due to the depletion of water in the pond decreases, and the new position of the i-th population individual is called It compares to the previous position and if there is a better value, it replaces it.

[0109] In the phase of increasing or decreasing the amount of water in the basin: the water in the pond will be restored by rain and snowmelt or spring water, the water will sink into the ground or holes, and even evaporation will cause a certain volume of water to evaporate. These two situations are mutually inverse. Assume that the probability of these two situations occurring (Pt )equal:

[0110]

[0111] This cycle continues until the required number of iterations is performed or an acceptable optimal solution is obtained.

[0112] In this control method, meteorological parameters are used as reference objects, and the flood optimization algorithm is used to find the optimal weights and thresholds of the BP neural network through continuous iterations, thereby obtaining better PID adjustment parameters.

[0113] Specifically, in this implementation, refer to Figure 6 and Figure 7 As shown, when training the neural network model, the weights and thresholds of the neural network model are updated using a flood optimization algorithm, and the flood optimization algorithm is set in MATLAB. The updating process of the weights and thresholds of the neural network model includes:

[0114] S01: Obtain the parameters of the flood optimization algorithm. Specifically, the parameters of the flood optimization algorithm include the maximum number of iterations Iter max , population scaling factor N e , population size N pop 、The current number of iterations Iter=0、The upper bound S max and the lower bound S min .

[0115] S02: Creating an initial population of a flood optimization algorithm according to the meteorological parameters, initial weight values, and initial threshold values;

[0116] S03: Determine the initial position for the start of iteration, and calculate the fitness of individuals in the initial population in each iteration, where the fitness is the performance of the neural network model under the current weights and thresholds.

[0117] Specifically, the fitness of an individual is calculated by the following formula:

[0118] S i =S min +rand×(S max -S min ) (6)

[0119] S04: updating the initial population through a flood optimization algorithm mechanism according to the fitness of the individuals in the initial population to obtain an updated population, wherein the individuals in the updated population are located at new positions;

[0120] Specifically, the position of the individual in the population before the update is the old position, and the position of the individual in the population after the update is the new position.

[0121] Specifically, step S04 includes:

[0122] S041: Calculate the individual optimization cost according to the fitness of the individuals in the initial population. Specifically, the individual optimization cost is calculated by formula (4).

[0123] S042: Determine the update type of the individual position based on the individual optimization cost, the random number of the current iteration and the random number of the previous iteration. Specifically, determine the update type of the individual position based on the random number of the current iteration and the random number of the previous iteration according to the following formula:

[0124] rand n >rand n-1 +P ei (7)

[0125] In the formula, rand n The random number for the current iteration; rand n-1 is the random number of the previous iteration.

[0126] When formula (7) is satisfied, the update type of the individual position is determined to be a fast flow mode.

[0127] When formula (7) is not satisfied, the update type of the individual position is determined to be a regular flow pattern.

[0128] S043: Based on the determined update type of the individual position, calculate the new position of the individual to obtain the updated population.

[0129] Specifically, when formula (7) is satisfied, formula (3) is selected to determine the update type of the individual position in a fast flow mode.

[0130] When formula (7) is not satisfied, formula (1) is selected to determine the update type of individual positions in a regular flow pattern.

[0131] S05: Based on the pre-set judgment rules, the position of the individual is updated.

[0132] Specifically, step S05 includes:

[0133] S051: Based on the preset optimization function, calculate the optimization function value when the individual is at the new position The optimization function value f(S i ); The preset optimization function can be set according to the model requirements to meet the iterative optimization requirements.

[0134] S052: When the optimization function value of the individual at the new position is less than the optimization function value of the individual at the old position, the position of the individual is determined by the old position S iReplace with new location And the optimization function value f(S best ) is replaced by the optimized function value when the individual is at the new position Otherwise, no action will be taken;

[0135] S053: The optimization function value f(S i ) is less than the optimization function value f(S best ), determine the position of the individual by the optimal position S best Replace with old position S i , and the optimization function value f(S best ) is replaced by the optimization function value f(S i ), otherwise, no processing is done.

[0136] Furthermore, when the population scaling factor N e When the value is between 1 and a preset value, the method for increasing or decreasing the water volume stage further includes:

[0137] S054: Calculate the probability of water volume increase or decrease based on the random number of the current iteration. Specifically, the probability of water volume increase or decrease is calculated by formula (5).

[0138] S055: Calculate the optimization function value when the individual is at the new position based on the comparison result of the random number of the current iteration and the probability of water volume increase or decrease. Specifically, if the random number rand of the current iteration is less than the probability P of water volume increase or decrease, t , that is, rand <P t , the new position of the individual is calculated by the following formula:

[0139]

[0140] Then, the optimization function value when the individual is in the new position is

[0141] S056: Optimization function value when the individual is in a new position is smaller than the optimal function value f(S best ), determine the position of the individual by the optimal position S best Replace with new location And the optimization function value f(S best ) is replaced by the optimized function value when the individual is at the new position

[0142]

[0143] S06: Based on the preset output conditions, the weight and threshold represented by the global optimal position are output.

[0144] Specifically, the preset output condition is that the number of iterations reaches the maximum number of iterations Iter max , output the weight and threshold represented by the global optimal position.

[0145] Specifically, as a specific embodiment, when training the neural network model, the weights and thresholds of the neural network model are updated using a flood optimization algorithm, and the updating process includes:

[0146] Step 1: Set the parameters of the flood optimization algorithm: the maximum number of iterations Iter max , population size (indicates the number of individuals in each iteration) N pop , population scaling factor N e 、The current number of iterations Iter=0、The upper bound S max and the lower bound S min ;

[0147] Step 2: Create the initial population of the flood optimization algorithm with meteorological parameters, initial weight values, and initial threshold values;

[0148] Step 3: Determine the initial position for the start of iteration, and use formula (6) to calculate the fitness of individuals in the initial population in each iteration;

[0149] Step 4: Calculate the individual optimization cost by (Formula 4);

[0150] Step 5: Determine whether formula (7) is valid;

[0151] Yes, in fast flow mode, select formula (3) to determine the update type of individual positions;

[0152] No, with a regular flow pattern, choose formula (1) to determine the update type of individual positions;

[0153] Step 6: Compare

[0154] Yes, the old and new positions are swapped f(S best )=f(S i );

[0155] Step 7: Compare f(S i ) <f(S best );

[0156] Yes, S. best =S i , f(S best )=f(S i );

[0157] Step 8: Calculate P using formula (5) t ;

[0158] Step 9: Determine rand <P t ;

[0159] Yes, the new position of the individual in the stage of increasing or decreasing water volume is calculated by formula (8):

[0160] Step 10: Compare

[0161] yes,

[0162] Step 11: If the population scaling factor N e When it is between 1 and the preset value, continue to loop steps 8 to 10;

[0163] Step 12: Determine whether the number of iterations reaches the maximum number of iterations Iter max ;

[0164] Step 13: No, continue to cycle steps 4 to 12;

[0165] Yes, return the value at the global best position, that is, the optimal weight and threshold of the BP neural network;

[0166] Finally, the BP neural network is used to obtain the PID adjustment parameters required for a better electric control valve.

[0167] Furthermore, the method further comprises:

[0168] Obtain fuel concentration detection value;

[0169] Based on the preset fuel concentration preset value, the start or stop of the explosion-proof fan 250 is controlled according to the fuel concentration detection value.

[0170] Furthermore, the method can be implemented through the MCGS configuration software monitoring system 120.

[0171] Further, the method comprises:

[0172] Based on user settings, the nitrogen valve 510 is controlled to open to purge the fuel pipeline.

[0173] Furthermore, the method further comprises:

[0174] In the high fire mode, the first solenoid valve 221, the second solenoid valve 222, the third solenoid valve 223, the fourth solenoid valve 224, the fifth solenoid valve 225, the sixth solenoid valve 226, the first electric regulating valve 211, the second electric regulating valve 212, the third electric regulating valve 213 and the fourth electric regulating valve 214 are controlled to open;

[0175] In the low fire mode, the first solenoid valve 221, the second solenoid valve 222, and the third solenoid valve 223 are controlled to be closed, and the fourth solenoid valve 224, the fifth solenoid valve 225, the sixth solenoid valve 226, the first electric regulating valve 211, the second electric regulating valve 212, the third electric regulating valve 213 and the fourth electric regulating valve 214 are opened, and the meteorological parameters are input as input values ​​into the pre-trained neural network model, and the output value of the pre-trained neural network model is the estimated opening of each regulating valve, and the opening of the corresponding regulating valve is adjusted according to the estimated opening of each regulating valve.

[0176] It can be seen that in the above scheme, the present application obtains the estimated value of the opening of the regulating valve by inputting the meteorological parameters into the pre-trained neural network model, and adjusts the regulating valve in the combustion system according to the estimated value of the opening of the regulating valve to control the intensity of the flame, thereby realizing the automatic adjustment of the torch combustion system, saving time and effort, and having high efficiency and good effect.

[0177] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0178] In one embodiment, a flare combustion control device based on meteorological parameters is provided, and the flare combustion control device based on meteorological parameters corresponds to the flare combustion control method based on meteorological parameters in the above embodiment. Fig. 9 As shown, the flare combustion control device based on meteorological parameters includes an acquisition module 101, a calculation module 102, and an adjustment module 103. The detailed description of each functional module is as follows:

[0179] The acquisition module 101 is used to acquire meteorological parameters, including at least one of temperature, air pressure, wind direction, wind force, light, and rainfall;

[0180] A calculation module 102, used for inputting the meteorological parameters as input values ​​into a pre-trained neural network model, wherein the output value of the pre-trained neural network model is an estimated opening degree of each regulating valve;

[0181] The adjustment module 103 is used to adjust the opening of the corresponding regulating valve according to the estimated opening of each regulating valve.

[0182] Specifically, the acquisition module 101 is also used to obtain the fuel concentration detection value, fuel pipeline pressure, nitrogen pressure, cabinet door start / stop, emergency stop button status, console start / stop, emergency stop button status, and flame status.

[0183] Specifically, the adjustment module 103 is further used to:

[0184] According to the estimated opening of the regulating valve and based on the mapping relationship between the opening of the regulating valve and the fuel pipeline pressure, a pipeline pressure setting value of the fuel pipeline where the regulating valve is located is obtained;

[0185] The opening of the regulating valve is adjusted according to the pipeline pressure setting value of the fuel pipeline where the regulating valve is located.

[0186] Specifically, the adjustment module 103 is further used to:

[0187] Based on the preset fuel concentration preset value, the start or stop of the explosion-proof fan is controlled according to the fuel concentration detection value.

[0188] Specifically, it also includes a training module 104, which is used to:

[0189] Get the parameters of the flood optimization algorithm;

[0190] Creating an initial population of a flood optimization algorithm according to the meteorological parameters, initial weight values, and initial threshold values;

[0191] Determine the initial position for the start of iteration, and calculate the fitness of individuals in the initial population in each iteration, where the fitness is the performance of the neural network model under the current weights and thresholds;

[0192] According to the fitness of individuals in the initial population, the initial population is updated by a flood optimization algorithm mechanism to obtain an updated population, wherein the individuals in the updated population are located at new positions;

[0193] Based on the pre-set judgment rules, the individual's position is updated;

[0194] Based on the preset output conditions, the weight and threshold represented by the global optimal position are output.

[0195] Specifically, the training module 104 is also used to:

[0196] Calculating individual optimization costs according to the fitness of individuals in the initial population;

[0197] Determining an update type of the individual position according to the individual optimization cost based on a random number of a current iteration and a random number of a previous iteration;

[0198] Based on the determined update type of the individual position, the new position of the individual is calculated to obtain the updated population.

[0199] Specifically, the training module 104 is also used to:

[0200] Based on the preset optimization function, the optimization function value when the individual is at the new position and the optimization function value when the individual is at the old position are calculated respectively;

[0201] When the optimization function value of the individual at the new position is less than the optimization function value of the individual at the old position, the position of the individual is determined to be replaced from the old position to the new position, and the optimization function value of the individual at the best position is replaced by the optimization function value of the individual at the new position, otherwise, no processing is performed;

[0202] When the optimization function value when the individual is at the old position is less than the optimization function value at the best position, the individual's position is determined to be replaced from the best position to the old position, and the optimization function value of the individual at the best position is replaced by the optimization function value when the individual is at the old position, otherwise, no processing is performed.

[0203] Specifically, the training module 104 is also used to:

[0204] Calculate the probability of water volume increasing or decreasing based on the random number of the current iteration;

[0205] According to the random number of the current iteration and the probability comparison result of water volume increase or decrease, the optimization function value when the individual is located at the new position is calculated;

[0206] When the optimization function value when the individual is at the new position is less than the optimization function value at the optimal position, it is determined that the position of the individual is replaced from the optimal position to the new position, and the optimization function value of the individual at the optimal position is replaced by the optimization function value when the individual is at the new position.

[0207] The present invention provides a torch combustion control device based on meteorological parameters. The device obtains an estimated value of the opening of a regulating valve by inputting the meteorological parameters into a pre-trained neural network model, and adjusts the regulating valve in the combustion system according to the estimated value of the opening of the regulating valve to control the intensity of the flame, thereby realizing automatic adjustment of the torch combustion system, saving time and labor, and having high efficiency and good effect.

[0208] For the specific definition of the flare combustion control device based on meteorological parameters, please refer to the definition of the flare combustion control method based on meteorological parameters in the above text, which will not be repeated here. Each module in the above-mentioned flare combustion control device based on meteorological parameters can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0209] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig.10As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external device end through a network connection. When the computer program is executed by the processor, it realizes the functions or steps of a flare combustion control method based on meteorological parameters on the service side.

[0210] In one embodiment, a computer device is provided. The computer device may be a device end, and its internal structure diagram may be as follows: Fig.11 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer program is executed by the processor, it realizes the functions or steps of a device side of a flare combustion control method based on meteorological parameters.

[0211] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0212] Obtaining meteorological parameters, including at least one of temperature, air pressure, wind direction, wind speed, light, and rainfall;

[0213] Inputting the meteorological parameters as input values ​​into a pre-trained neural network model, the output value of the pre-trained neural network model being the estimated opening of each regulating valve;

[0214] The opening of the corresponding regulating valve is adjusted according to the estimated opening of each regulating valve.

[0215] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0216] Obtaining meteorological parameters, including at least one of temperature, air pressure, wind direction, wind speed, light, and rainfall;

[0217] Inputting the meteorological parameters as input values ​​into a pre-trained neural network model, the output value of the pre-trained neural network model being the estimated opening of each regulating valve;

[0218] The opening of the corresponding regulating valve is adjusted according to the estimated opening of each regulating valve.

[0219] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the device side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0220] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0221] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0222] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A flare combustion control method based on meteorological parameters, characterized in that: include: Obtaining meteorological parameters, including at least one of temperature, air pressure, wind direction, wind speed, light, and rainfall; Inputting the meteorological parameters as input values ​​into a pre-trained neural network model, the output value of the pre-trained neural network model being the estimated opening of each regulating valve; The opening of the corresponding regulating valve is adjusted according to the estimated opening of each regulating valve.

2. The method according to claim 1, characterized in that: The step of adjusting the corresponding regulating valve according to the estimated opening of each regulating valve comprises: According to the estimated opening of the regulating valve and based on the mapping relationship between the opening of the regulating valve and the fuel pipeline pressure, a pipeline pressure setting value of the fuel pipeline where the regulating valve is located is obtained; The opening of the regulating valve is adjusted according to the pipeline pressure setting value of the fuel pipeline where the regulating valve is located.

3. The method according to claim 1, characterized in that The method further comprises: When training the neural network model, the weights and thresholds of the neural network model are updated using the flood optimization algorithm, and the updating process includes: Get the parameters of the flood optimization algorithm; Creating an initial population of a flood optimization algorithm according to the meteorological parameters, initial weight values, and initial threshold values; Determine the initial position for the start of iteration, and calculate the fitness of individuals in the initial population in each iteration, where the fitness is the performance of the neural network model under the current weights and thresholds; According to the fitness of individuals in the initial population, the initial population is updated by a flood optimization algorithm mechanism to obtain an updated population, wherein the individuals in the updated population are located at new positions; Based on the pre-set judgment rules, the individual's position is updated; Based on the preset output conditions, the weight and threshold represented by the global optimal position are output.

4. The method according to claim 3, characterized in that According to the fitness of individuals in the initial population, the initial population is updated by a flood optimization algorithm mechanism to obtain an updated population, where individuals in the updated population are located at new positions, including: Calculating individual optimization costs according to the fitness of individuals in the initial population; Determining an update type of the individual position according to the individual optimization cost based on a random number of a current iteration and a random number of a previous iteration; Based on the determined update type of the individual position, the new position of the individual is calculated to obtain the updated population.

5. The method according to claim 3, characterized in that: Based on the pre-set judgment rules, the individual's location is updated, including: Based on the preset optimization function, the optimization function value when the individual is at the new position and the optimization function value when the individual is at the old position are calculated respectively; When the optimization function value when the individual is at the new position is less than the optimization function value when the individual is at the old position, it is determined that the position of the individual is replaced from the old position to the new position, and the optimization function value when the individual is at the best position is replaced by the optimization function value when the individual is at the new position; When the optimization function value when the individual is at the old position is less than the optimization function value at the optimal position, it is determined that the position of the individual is replaced from the optimal position to the old position, and the optimization function value of the individual at the optimal position is replaced by the optimization function value when the individual is at the old position.

6. The method according to claim 3, characterized in that Based on the pre-set judgment rules, the individual's location is updated, including: Calculate the probability of water volume increasing or decreasing based on the random number of the current iteration; According to the random number of the current iteration and the probability comparison result of water volume increase or decrease, the optimization function value when the individual is located at the new position is calculated; When the optimization function value when the individual is at the new position is less than the optimization function value at the optimal position, it is determined that the position of the individual is replaced from the optimal position to the new position, and the optimization function value of the individual at the optimal position is replaced by the optimization function value when the individual is at the new position.

7. The method according to claim 1, characterized in that The method further comprises: Obtain fuel concentration detection value; Based on the preset fuel concentration preset value, the start or stop of the explosion-proof fan is controlled according to the fuel concentration detection value.

8. A flare combustion control device based on meteorological parameters, characterized in that: include: An acquisition module, used to acquire meteorological parameters, including at least one of temperature, air pressure, wind direction, wind force, light, and rainfall; A calculation module, used for inputting the meteorological parameters as input values ​​into a pre-trained neural network model, wherein the output value of the pre-trained neural network model is an estimated opening degree of each regulating valve; The adjustment module is used to adjust the opening of the corresponding regulating valve according to the estimated opening of each regulating valve.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the flare combustion control method based on meteorological parameters as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the flare combustion control method based on meteorological parameters as claimed in any one of claims 1 to 7 are implemented.

Images