Intelligent Temperature Control Method for Hot Blast Stove Based on Working Condition Recognition Technology and Fuzzy PID

Abstract

The present invention provides a method for intelligent temperature control of a hot blast stove based on operating condition recognition technology and fuzzy PID, comprising the following steps: S1: Collect historical operation data of the hot blast stove; S2: Construct a multi-variable operating condition matrix; S3: Establish a multi-condition parameter self-tuning model based on the multi-variable operating condition matrix; S4: Real-time identify the target temperature under the current operating condition; S5: Design a control system to regulate the current furnace temperature, post-furnace temperature, and SCR reactor inlet temperature; The method for intelligent temperature control of a hot blast stove based on operating condition recognition technology and fuzzy PID proposed by the present invention realizes adaptive regulation of the hot blast stove temperature under different operating conditions, can improve the robustness and adaptability of the system, and enables the hot blast stove to maintain stable temperature control under different operating states or environmental conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of furnace temperature control, and particularly to a method for intelligent temperature control of a hot blast stove based on working condition recognition technology and fuzzy PID. Background Art

[0002] Heating the gas in the denitrification system by the hot blast stove is an important link in the SCR denitrification process and directly affects the denitrification efficiency in the SCR reactor; if the flue gas temperature at the reactor inlet < 300 °C, the denitrification efficiency will decrease due to the low inlet flue gas temperature, resulting in an increase in ammonia consumption; however, if the flue gas temperature at the reactor inlet is too high, the high-temperature inlet gas will accelerate the evaporation rate of ammonia in the ammonia water, leading to an increase in the evaporation of ammonia water, and the high-temperature gas environment will not only increase the consumption of blast furnace gas at the hot blast stove, but also increase the rate of ammonia escaping from the ammonia water into the air, greatly increasing the probability of ammonia escape, reducing the amount of ammonia available for the denitrification reaction, and possibly causing the SCR system to fail to achieve the expected denitrification effect; the high-temperature gas environment may also have an adverse effect on the activity of the SCR catalyst, and the evaporation of ammonia water and ammonia escape will lead to an uneven distribution of ammonia concentration, thereby affecting the uniformity of the SCR reaction; it may cause nitrogen oxides (NOx) in some areas to not be effectively reduced, reducing the overall denitrification efficiency.

[0003] Currently, PID control technology is widely used in the intelligent temperature control system of the hot blast stove. By adjusting three key control factors, namely the gas valve, the combustion-supporting air valve, and the air blending valve, the furnace temperature, the temperature after the furnace, and the temperature at the inlet of the SCR reactor are controlled in the ideal working range in real time, thereby improving the denitrification efficiency of the SCR reactor, reducing ammonia water consumption, saving the use of blast furnace gas and combustion-supporting air, and achieving the ideal effect of cost reduction and efficiency increase; however, in the actual operation process, there are two problems in the intelligent temperature control system of the hot blast stove:

[0004] Lack of working condition recognition: In the hot blast stove system, there are complex interrelationships among parameters such as the furnace temperature, the temperature after the furnace, and the temperature at the inlet of the SCR reactor under different working conditions; the system does not fully combine auxiliary factors such as the gas pipeline pressure, the catalyst reaction efficiency, and the ammonia water evaporation efficiency for comprehensive analysis and recognition of the working conditions, and it is difficult to intelligently adjust the target values of parameters such as the furnace temperature, the temperature after the furnace, and the temperature at the inlet of the SCR reactor according to different working conditions, thereby affecting the overall denitrification effect;

[0005] Slow reaction speed and difficult parameter adjustment; during the temperature adjustment process of the hot blast stove, the reaction speed is relatively slow. Especially in the aspect of PID parameter adjustment, a large amount of manual experience comparison and on-site attempts are required. This makes the system unable to make timely adjustments when facing rapidly changing working conditions. Moreover, the hot blast stove system has strong coupling, and the key control factors affect each other, which makes the PID parameter adjustment under different working conditions more difficult, unable to flexibly and independently adjust each parameter, and reduces the stability and controllability of the entire system. Summary of the Invention

[0006] The purpose of the present invention is to solve the deficiencies existing in the prior art, and a hot blast stove temperature intelligent control method based on working condition recognition technology and fuzzy PID is proposed.

[0007] To achieve the above purpose, the present invention adopts the following technical solutions:

[0008] A hot blast stove temperature intelligent control method based on working condition recognition technology and fuzzy PID includes the following steps:

[0009] S1: Collect historical operation data of the hot blast stove;

[0010] Collect the historical operation data of the hot blast stove and the hot blast stove temperature under different working conditions;

[0011] The hot blast stove temperature includes the actual value of the hot blast stove temperature and the target value of the hot blast stove temperature;

[0012] The historical operation data includes the actual gas pressure, the gas flow rate of the first-stage hot blast stove, the combustion-supporting air flow rate, the actual opening degree of the gas valve, the actual opening degree of the blast valve, the actual opening degree of the air mixing valve, the inlet oxygen content, etc.;

[0013] The actual value of the hot blast stove temperature includes the actual value of the furnace temperature, the actual value of the temperature after the furnace, and the actual value of the temperature at the inlet of the SCR reactor;

[0014] The target value of the hot blast stove temperature includes the target value of the furnace temperature, the target value of the temperature after the furnace, and the target value of the temperature at the inlet of the SCR reactor;

[0015] The different working conditions refer to different operating states or environmental conditions, such as different loads, different burners, different fuel types, etc.

[0016] S2: Construct a multi-variable working condition matrix;

[0017] Construct a multi-variable working condition matrix for hot blast stove temperature control and optimization according to the collected historical operation data of the hot blast stove; the rows of the multi-variable working condition matrix correspond to different working conditions, and the columns correspond to each historical operation data.

[0018] S3: Establish a multi-condition parameter self-tuning model according to the multi-variable working condition matrix;

[0019] Based on the variable operating condition matrix, a multi-condition parameter self-tuning model for the hot blast stove temperature control is established. The establishment of the multi-condition parameter self-tuning model includes the following steps:

[0020] S31: Feature parameter selection;

[0021] Select the historical operation data related to the hot blast stove temperature from the historical operation data, and extract the features of the historical operation data;

[0022] The features include average value, variance, maximum value, minimum value, etc.;

[0023] S32: Construct an association model;

[0024] Based on the data relationship between the extracted features and the hot blast stove temperature target value collected in step S1, construct an association model between the features and the hot blast stove temperature target value, denoted as the multi-condition parameter self-tuning model;

[0025] S33: Perform parameter tuning;

[0026] Input the historical operation data under different operating conditions in the variable operating condition matrix into the multi-condition parameter self-tuning model, output the predicted hot blast stove temperature target value, compare the difference between the output predicted hot blast stove temperature target value and the actually collected hot blast stove temperature target value, and tune the parameters in the multi-condition parameter self-tuning model through the parameter optimization algorithm.

[0027] S4: Real-time identify the target temperature under the current operating condition;

[0028] A variety of sensors are arranged inside the hot blast stove;

[0029] Collect the operation data under the current operating condition through the sensors in real time, input the operation data under the current operating condition into the multi-condition parameter self-tuning model, and the multi-condition parameter self-tuning model identifies the furnace temperature target value, the temperature target value behind the furnace, and the SCR reactor inlet temperature target value under the current operating condition according to the operation data;

[0030] The operation data includes the actual gas pressure, the gas flow rate of the first-stage hot blast stove, the combustion-supporting air flow rate, the actual opening of the gas valve, the actual opening of the blast valve, the actual opening of the air mixing valve, the inlet oxygen content, etc.

[0031] S5: Design a control system to regulate the current furnace temperature, the temperature behind the furnace, and the SCR reactor inlet temperature;

[0032] The control system includes a fuzzy cascade control system for the reactor inlet temperature, a fuzzy variable ratio control system for the furnace temperature, and a fuzzy PID control system for the temperature behind the furnace;

[0033] The control system combines the identified furnace temperature target value, post-furnace temperature target value, and SCR reactor inlet temperature target value in step S4 to control the current furnace temperature, post-furnace temperature, and SCR reactor inlet temperature respectively;

[0034] Specifically, it includes the following sub-steps:

[0035] S51: The reactor inlet temperature fuzzy cascade control system regulates the SCR reactor inlet temperature;

[0036] The main loop of the reactor inlet temperature fuzzy-cascade control system includes a reactor temperature fuzzy controller, a gas flow controller, and a gas valve actuator;

[0037] Specifically, it includes the following sub-steps:

[0038] S511: Calculate the temperature error and error derivative between the actual value and the target value of the reactor inlet temperature;

[0039] A temperature sensor is internally installed at the reactor inlet, and the real-time actual value of the reactor inlet temperature is collected through the temperature sensor;

[0040] The temperature error is obtained based on the actual value of the reactor inlet temperature and the reactor inlet temperature target value identified in step S4. The temperature error is the difference between the reactor inlet temperature target value and the actual furnace temperature;

[0041] Furthermore, the differential algorithm calculates the error derivative based on the actual values of the reactor inlet temperature at each timestamp in the previous period collected by the temperature sensor and the reactor inlet temperature target values at each timestamp in the previous period identified by the multi-condition parameter self-tuning model;

[0042] S512: Obtain the gas flow error and error derivative;

[0043] Input the temperature error and error derivative obtained in step S511 into the reactor temperature fuzzy controller, and the reactor temperature fuzzy controller outputs the gas flow;

[0044] Furthermore, the gas flow at each timestamp in the previous period is obtained through the reactor temperature fuzzy controller;

[0045] The sensor collects the current actual gas flow in real time;

[0046] The differential algorithm obtains the gas flow error based on the output gas flow and the current actual gas flow, and obtains the gas flow error derivative based on the actual gas flows at each timestamp in the previous period collected by the sensor and the gas flows at each timestamp in the previous period obtained by the reactor temperature fuzzy controller;

[0047] S513: Regulate the reactor inlet temperature;

[0048] Input the gas flow error and the derivative of the gas flow error into the gas flow controller. The gas flow controller adjusts the opening degree of the gas valve according to fuzzy rules. The gas flow controller outputs the adjusted opening degree of the gas valve, and inputs the valve opening degree into the gas valve actuator to obtain the actual value of the reactor inlet temperature in the next control cycle;

[0049] Loop steps S511 - S513 to regulate the reactor inlet temperature until the actual value of the reactor inlet temperature approximates the target value of the reactor inlet temperature.

[0050] Further, take the gas pressure as the secondary loop and offset the change in gas pressure by adjusting the gas flow;

[0051] S52: The furnace temperature fuzzy PID control system regulates the furnace temperature;

[0052] The furnace temperature fuzzy PID control system includes a temperature fuzzy controller, a temperature fuzzy compensation controller, a flow controller, and a combustion air valve actuator;

[0053] Specifically, it includes the following sub - steps:

[0054] S521: Calculate the temperature error and the derivative of the error between the actual value and the target value of the furnace temperature;

[0055] A temperature sensor is installed inside the furnace, and the real - time actual value of the furnace temperature is collected through the temperature sensor;

[0056] Obtain the temperature error according to the actual value of the furnace temperature and the target value of the furnace temperature identified in step S4. The temperature error is the difference between the target value of the furnace temperature and the actual value of the furnace temperature;

[0057] Further, the differential algorithm calculates the derivative of the error based on the actual values of the furnace temperature at each timestamp in the previous period collected by the temperature sensor and the target values of the furnace temperature at each timestamp in the previous period identified by the multi - operating - condition parameter self - tuning model;

[0058] S522: Obtain the current air - fuel ratio set value;

[0059] Input the temperature error and the derivative of the error into the temperature fuzzy controller, and the temperature fuzzy controller outputs the current air - fuel ratio set value;

[0060] S523: Output the opening degree of the combustion air valve in the current control cycle;

[0061] Calculate the actual value of the combustion air flow according to the air - fuel ratio and the gas flow. The actual value of the combustion air flow = air - fuel ratio * gas flow;

[0062] The air-fuel ratio is the current air-fuel ratio set value obtained in step S522, and the gas flow rate is obtained by real-time collection by a sensor in the furnace;

[0063] Further, the furnace temperature fuzzy PID control system preset a target value of the combustion-supporting air flow rate in advance. According to the target value of the combustion-supporting air flow rate and the actual value of the combustion-supporting air flow rate, the error of the combustion-supporting air flow rate is obtained. The error of the combustion-supporting air flow rate is the difference between the target value of the combustion-supporting air flow rate and the actual value of the combustion-supporting air flow rate;

[0064] Further, according to the gas flow rate at each time stamp in the previous period collected by the sensor, the actual value of the combustion-supporting air flow rate at each time stamp in the previous period is obtained;

[0065] Further, the differential algorithm calculates the derivative of the combustion-supporting air flow rate error according to the actual value of the combustion-supporting air flow rate at each time stamp in the previous period and the target value of the combustion-supporting air flow rate preset by the furnace temperature fuzzy PID control system;

[0066] Input the combustion-supporting air flow rate error and the derivative of the combustion-supporting air flow rate error into the combustion-supporting air flow rate controller. The combustion-supporting air flow rate controller outputs the opening degree of the combustion-supporting air valve in the current control period, denoted as the opening degree u2 of the combustion-supporting air valve;

[0067] S524: Regulate the furnace temperature;

[0068] Preferably, it further includes a feedforward temperature fuzzy compensation controller. Input the temperature error and the derivative of the error of the furnace temperature obtained in step S521 into the temperature fuzzy compensation controller. Adjust the opening degree of the combustion-supporting air valve according to the fuzzy rules, and output the adjusted opening degree of the combustion-supporting air valve, denoted as the opening degree du of the combustion-supporting air valve;

[0069] Obtain the actual opening degree of the combustion-supporting air valve. The actual opening degree of the combustion-supporting air valve = the opening degree u2 of the combustion-supporting air valve + the opening degree du of the combustion-supporting air valve; Input the actual opening degree of the combustion-supporting air valve into the combustion-supporting air valve actuator to obtain the actual value of the furnace temperature in the next control period;

[0070] Loop steps S521 - S524 to adjust the furnace temperature until the actual value of the furnace temperature is approximately equal to the target value of the furnace temperature;

[0071] S53: The post-furnace temperature fuzzy variable ratio control system regulates the post-furnace temperature;

[0072] The post-furnace temperature fuzzy variable ratio control system includes a post-furnace temperature controller and an air mixing valve actuator;

[0073] It includes the following sub-steps:

[0074] S531: Calculate the temperature error and the derivative of the error between the actual value and the target value of the post-furnace temperature;

[0075] A temperature sensor is installed inside the hot blast stove, and the actual value of the temperature behind the stove is collected in real time through the temperature sensor;

[0076] A temperature error is obtained based on the actual value of the temperature behind the stove and the target value of the temperature behind the stove identified in step S4. The temperature error is the difference between the target value of the temperature behind the stove and the actual value of the temperature behind the stove;

[0077] Furthermore, the differential algorithm calculates the derivative of the temperature error behind the stove based on the actual values of the temperature behind the stove at each time stamp in the previous period collected by the temperature sensor and the target values of the temperature behind the stove at each time stamp in the previous period identified by the multi-condition parameter self-tuning model;

[0078] S532: Obtain the opening degree of the air mixing valve;

[0079] Input the temperature error behind the stove and the derivative of the temperature error behind the stove into the temperature controller behind the stove. The temperature controller behind the stove adjusts the opening degree of the air mixing valve according to the fuzzy rules and outputs the adjusted opening degree of the air mixing valve;

[0080] Input the output opening degree of the air mixing valve into the air mixing valve actuator, and the air mixing valve actuator outputs the actual value of the temperature behind the stove in the next control cycle;

[0081] Loop steps S531 - S532 to adjust the furnace temperature until the actual value of the furnace temperature is approximately equal to the target value of the furnace temperature;

[0082] Furthermore, the fuzzy rules are as follows:

[0083] If the actual value of the temperature behind the stove is lower than the target value of the temperature behind the stove and the deviation continues to increase, increase the opening degree of the air mixing valve;

[0084] If the actual value of the temperature behind the stove is lower than the target value of the temperature behind the stove but the deviation gradually decreases, slightly increase the opening degree of the air mixing valve;

[0085] If the actual value of the temperature behind the stove is higher than the target value of the temperature behind the stove and the deviation continues to increase, decrease the opening degree of the air mixing valve;

[0086] If the actual value of the temperature behind the stove is higher than the target value of the temperature behind the stove but the deviation gradually decreases, slightly decrease the opening degree of the air mixing valve;

[0087] If the actual value of the temperature behind the stove is close to the target value of the temperature behind the stove, keep the current opening degree of the air mixing valve unchanged;

[0088] If the actual value of the temperature behind the stove is far from the target value of the temperature behind the stove and the deviation is large, greatly adjust the opening degree of the air mixing valve.

[0089] Compared with the prior art, the beneficial effects of the present invention are:

[0090] The intelligent temperature control method of the hot blast stove based on the working condition recognition technology and fuzzy PID proposed by the present invention analyzes historical data and recognizes working conditions, and comprehensively analyzes and recognizes working conditions by combining auxiliary factors such as gas pipeline pressure, catalyst reaction efficiency, and ammonia evaporation efficiency to obtain the target values of the furnace temperature, post-furnace temperature, and SCR reactor inlet temperature under different working conditions. Then, a fuzzy variable ratio furnace temperature control system, a fuzzy PID post-furnace temperature control system, and a fuzzy cascade reactor inlet temperature control system are designed. The controller is tuned through fuzzy rules, and three key control factors, namely the opening degree of the gas valve, the opening degree of the combustion-supporting air valve, and the opening degree of the air blending valve, are adjusted to control the furnace temperature, post-furnace temperature, and SCR reactor inlet temperature to be stably near the target temperature in real time;

[0091] This method realizes the adaptive regulation of the hot blast stove temperature under different working conditions, can improve the robustness and adaptability of the system, and enables the hot blast stove to maintain stable temperature control under different operating states or environmental conditions;

[0092] The intelligent temperature regulation of the hot blast stove is carried out through the designed control system. The temperature regulation speed is fast, and there is no need for manual comparison or on-site trial. When facing rapidly changing working conditions, adjustments can be made in a timely manner. By tuning the parameters through fuzzy rules, various parameters can be flexibly called to improve the stability and controllability of the system. Brief Description of the Drawings

[0093] Figure 1 It is the step flow chart of the intelligent temperature control method of the hot blast stove based on the working condition recognition technology and fuzzy PID of the present invention;

[0094] Figure 2 It is the fuzzy cascade control structure diagram of the reactor inlet temperature of the intelligent temperature control method of the hot blast stove based on the working condition recognition technology and fuzzy PID of the present invention;

[0095] Figure 3 It is the structure diagram of the fuzzy variable ratio control system of the furnace temperature of the intelligent temperature control method of the hot blast stove based on the working condition recognition technology and fuzzy PID of the present invention;

[0096] Figure 4 It is the structure diagram of the fuzzy variable ratio control system of the post-furnace temperature of the intelligent temperature control method of the hot blast stove based on the working condition recognition technology and fuzzy PID of the present invention; Detailed Embodiment

[0097] To further understand the purpose, structure, features, and functions of the present invention, the following is a detailed description in conjunction with embodiments.

[0098] As shown in Figure 1 、 2 、3, and 4, the intelligent temperature control method of the hot blast stove based on the working condition recognition technology and fuzzy PID includes the following steps:

[0099] S1: Collect historical operation data of the hot blast stove;

[0100] Collect historical operation data of the hot blast stove and the temperature of the hot blast stove under different working conditions;

[0101] The temperature of the hot blast stove includes the actual value of the hot blast stove temperature and the target value of the hot blast stove temperature;

[0102] The historical operation data includes the actual gas pressure, the gas flow rate of the first-stage hot blast stove, the combustion-supporting air flow rate, the actual opening degree of the gas valve, the actual opening degree of the blast valve, the actual opening degree of the air blending valve, the inlet oxygen content, etc.;

[0103] The actual value of the hot blast stove temperature includes the actual value of the furnace temperature, the actual value of the temperature behind the furnace, and the actual value of the temperature at the inlet of the SCR reactor;

[0104] The target value of the hot blast stove temperature includes the target value of the furnace temperature, the target value of the temperature behind the furnace, and the target value of the temperature at the inlet of the SCR reactor;

[0105] The different working conditions refer to different operating states or environmental conditions, such as different loads, different burners, different fuel types, etc.

[0106] S2: Construct a multi-variable working condition matrix;

[0107] Construct a multi-variable working condition matrix for the temperature control and optimization of the hot blast stove based on the collected historical operation data of the hot blast stove; the rows of the multi-variable working condition matrix correspond to different working conditions, and the columns correspond to each historical operation data.

[0108] S3: Establish a multi-condition parameter self-tuning model according to the multi-variable working condition matrix;

[0109] Establish a multi-condition parameter self-tuning model for the temperature control of the hot blast stove according to the multi-variable working condition matrix. The establishment of the multi-condition parameter self-tuning model includes the following steps:

[0110] S31: Feature parameter selection;

[0111] Select the historical operation data related to the hot blast stove temperature from the historical operation data, and extract the features of the historical operation data;

[0112] The features include the average value, variance, maximum value, minimum value, etc.;

[0113] S32: Construct a correlation model;

[0114] Based on the relationship between the extracted features and the data of the target value of the hot blast stove temperature collected in step S1, construct a correlation model between the features and the target value of the hot blast stove temperature, denoted as the multi-condition parameter self-tuning model;

[0115] S33: Perform parameter tuning;

[0116] Input the historical operation data under different working conditions in the multi-condition matrix into the multi-condition parameter self-tuning model, output the predicted temperature target value of the hot blast stove, compare the difference between the output predicted temperature target value of the hot blast stove and the actually collected temperature target value of the hot blast stove, and tune the parameters in the multi-condition parameter self-tuning model through the parameter optimization algorithm.

[0117] S4: Real-time identify the target temperature under the current working condition;

[0118] A variety of sensors are arranged inside the hot blast stove;

[0119] Collect the operation data under the current working condition in real time through the sensors, input the operation data under the current working condition into the multi-condition parameter self-tuning model, and the multi-condition parameter self-tuning model identifies the target value of the furnace temperature, the target value of the temperature after the furnace, and the target value of the temperature at the inlet of the SCR reactor under the current working condition according to the operation data;

[0120] The operation data includes the actual gas pressure, the gas flow rate of the first-stage hot blast stove, the combustion-supporting air flow rate, the actual opening of the gas valve, the actual opening of the blast valve, the actual opening of the air blending valve, the inlet oxygen content, etc.

[0121] S5: Design a control system to regulate the current furnace temperature, temperature after the furnace, and temperature at the inlet of the SCR reactor;

[0122] The control system includes a fuzzy cascade control system for the temperature at the inlet of the reactor, a fuzzy variable ratio control system for the furnace temperature, and a fuzzy PID control system for the temperature after the furnace;

[0123] Control the current furnace temperature, temperature after the furnace, and temperature at the inlet of the SCR reactor respectively through the control system combined with the target values of the furnace temperature, temperature after the furnace, and temperature at the inlet of the SCR reactor identified in step S4;

[0124] Specifically, it includes the following sub-steps:

[0125] S51: The fuzzy cascade control system for the temperature at the inlet of the reactor regulates the temperature at the inlet of the SCR reactor;

[0126] The main loop of the fuzzy cascade control system for the temperature at the inlet of the reactor includes a reactor temperature fuzzy controller, a gas flow controller, and a gas valve actuator;

[0127] Specifically, it includes the following sub-steps:

[0128] S511: Calculate the temperature error and error derivative between the actual value and the target value of the temperature at the inlet of the reactor;

[0129] A temperature sensor is installed inside the reactor inlet, and the actual value of the reactor inlet temperature in real time is collected through the temperature sensor;

[0130] A temperature error is obtained based on the actual value of the reactor inlet temperature and the target value of the reactor inlet temperature identified in step S4. The temperature error is the difference between the target value of the reactor inlet temperature and the actual value of the furnace temperature;

[0131] Furthermore, the differential algorithm calculates the error derivative based on the actual values of the reactor inlet temperature at each time stamp in the previous period collected by the temperature sensor and the target values of the reactor inlet temperature at each time stamp in the previous period identified by the multi-condition parameter self-tuning model;

[0132] S512: Obtain the gas flow error and the error derivative;

[0133] The temperature error and the error derivative obtained in step S511 are input into the reactor temperature fuzzy controller, and the reactor temperature fuzzy controller outputs the gas flow;

[0134] Furthermore, the gas flow at each time stamp in the previous period is obtained through the reactor temperature fuzzy controller;

[0135] The sensor collects the current actual gas flow in real time;

[0136] The differential algorithm obtains the gas flow error based on the output gas flow and the current actual gas flow, and obtains the gas flow error derivative based on the actual gas flow at each time stamp in the previous period collected by the sensor and the gas flow at each time stamp in the previous period obtained by the reactor temperature fuzzy controller;

[0137] S513: Regulate the reactor inlet temperature;

[0138] The gas flow error and the gas flow error derivative are input into the gas flow controller. The gas flow controller adjusts the opening of the gas valve according to the fuzzy rules. The gas flow controller outputs the adjusted opening of the gas valve, and the valve opening is input into the gas valve actuator to obtain the actual value of the reactor inlet temperature in the next control cycle;

[0139] Loop steps S511 - S513 to regulate the reactor inlet temperature until the actual value of the reactor inlet temperature is approximately equal to the target value of the reactor inlet temperature.

[0140] Furthermore, the gas pressure is used as the secondary loop to offset the change in gas pressure by adjusting the gas flow;

[0141] Due to the change in gas pressure, the gas flow rate will also change, and the gas pressure changes frequently and is difficult to predict. Taking the gas pressure as the secondary loop, the gas flow rate is adjusted to offset the change in gas pressure, reducing the impact of gas pressure on the main loop (reactor inlet temperature);

[0142] S52: The furnace temperature fuzzy PID control system regulates the furnace temperature;

[0143] The furnace temperature fuzzy PID control system includes a temperature fuzzy controller, a temperature fuzzy compensation controller, a flow controller, and a combustion air valve actuator;

[0144] Specifically, it includes the following sub-steps:

[0145] S521: Calculate the temperature error and error derivative between the actual furnace temperature value and the target value;

[0146] A temperature sensor is installed inside the furnace, and the real-time actual furnace temperature value is collected through the temperature sensor;

[0147] The temperature error is obtained based on the actual furnace temperature value and the furnace temperature target value identified in step S4. The temperature error is the difference between the furnace temperature target value and the actual furnace temperature value;

[0148] Furthermore, the differential algorithm calculates the error derivative based on the actual furnace temperature values at each timestamp in the previous period collected by the temperature sensor and the furnace temperature target values at each timestamp in the previous period identified by the multi-condition parameter self-tuning model;

[0149] S522: Obtain the current air-fuel ratio set value;

[0150] The temperature error and error derivative are input into the temperature fuzzy controller, and the temperature fuzzy controller outputs the current air-fuel ratio set value;

[0151] S523: Output the opening degree of the combustion air valve in the current control cycle;

[0152] Calculate the actual combustion air flow rate based on the air-fuel ratio and the gas flow rate. The actual combustion air flow rate = air-fuel ratio * gas flow rate;

[0153] The air-fuel ratio is the current air-fuel ratio set value obtained in step S522, and the gas flow rate is obtained by real-time collection by the sensor inside the furnace;

[0154] Furthermore, the furnace temperature fuzzy PID control system has a pre-set target value for the combustion air flow rate. Based on the target value of the combustion air flow rate and the actual combustion air flow rate, the combustion air flow rate error is obtained. The combustion air flow rate error is the difference between the target value of the combustion air flow rate and the actual combustion air flow rate;

[0155] Further, based on the gas flow rates at each timestamp in the previous period collected by the sensor, the actual values of the combustion-supporting air flow rates at each timestamp in the previous period are obtained;

[0156] Further, the differential algorithm calculates the derivative of the combustion-supporting air flow rate error based on the actual values of the combustion-supporting air flow rates at each timestamp in the previous period and the target value of the combustion-supporting air flow rate preset by the furnace temperature fuzzy PID control system;

[0157] The combustion-supporting air flow rate error and the derivative of the combustion-supporting air flow rate error are input into the combustion-supporting air flow rate controller, and the combustion-supporting air valve opening degree in the current control cycle is output by the combustion-supporting air flow rate controller, denoted as the combustion-supporting air valve opening degree u2;

[0158] S524: Regulate the furnace temperature;

[0159] Preferably, it further includes a feedforward temperature fuzzy compensation controller. The temperature error and the derivative of the error of the furnace temperature obtained in step S521 are input into the temperature fuzzy compensation controller. According to the fuzzy rules, the combustion-supporting air valve opening degree is adjusted, and the adjusted combustion-supporting air valve opening degree is output, denoted as the combustion-supporting air valve opening degree du;

[0160] The actual combustion-supporting air valve opening degree is obtained, and the actual combustion-supporting air valve opening degree = combustion-supporting air valve opening degree u2 + combustion-supporting air valve opening degree du; The actual combustion-supporting air valve opening degree is input into the combustion-supporting valve actuator to obtain the actual value of the furnace temperature in the next control cycle;

[0161] Loop steps S521 - S524 to adjust the furnace temperature until the actual value of the furnace temperature is approximately equal to the target value of the furnace temperature;

[0162] S53: The post-furnace temperature fuzzy variable ratio control system regulates the post-furnace temperature;

[0163] The post-furnace temperature fuzzy variable ratio control system includes a post-furnace temperature controller and an air mixing valve actuator;

[0164] It includes the following sub-steps:

[0165] S531: Calculate the temperature error and the derivative of the error between the actual value and the target value of the post-furnace temperature;

[0166] A temperature sensor is installed inside the hot blast stove, and the real-time actual value of the post-furnace temperature is collected through the temperature sensor;

[0167] The temperature error is obtained according to the actual value of the post-furnace temperature and the target value of the post-furnace temperature identified in step S4, and the temperature error is the difference between the target value of the post-furnace temperature and the actual value of the post-furnace temperature;

[0168] Further, the differential algorithm calculates the derivative of the post-furnace temperature error based on the actual post-furnace temperature values at each timestamp in the previous period collected by the temperature sensor and the target post-furnace temperature values at each timestamp in the previous period identified by the multi-condition parameter self-tuning model;

[0169] S532: Obtain the opening degree of the air mixing valve;

[0170] Input the post-furnace temperature error and the derivative of the post-furnace temperature error into the post-furnace temperature controller. The post-furnace temperature controller adjusts the opening degree of the air mixing valve according to the fuzzy rules and outputs the adjusted opening degree of the air mixing valve;

[0171] Input the output opening degree of the air mixing valve into the air mixing valve actuator, and the air mixing valve actuator outputs the actual post-furnace temperature value for the next control cycle;

[0172] Loop steps S531 - S532 to adjust the furnace temperature until the actual furnace temperature value is approximately equal to the target furnace temperature value;

[0173] Further, the fuzzy rules are as follows:

[0174] If the actual hot blast stove temperature is lower than the target hot blast stove temperature and the deviation continues to increase, increase the opening degrees of the air mixing valve, the combustion-supporting air valve, and the gas valve;

[0175] If the actual hot blast stove temperature is lower than the target hot blast stove temperature but the deviation gradually decreases, slightly increase the opening degrees of the air mixing valve, the combustion-supporting air valve, and the gas valve;

[0176] If the actual hot blast stove temperature is higher than the target hot blast stove temperature and the deviation continues to increase, decrease the opening degrees of the air mixing valve, the combustion-supporting air valve, and the gas valve;

[0177] If the actual hot blast stove temperature is higher than the target hot blast stove temperature but the deviation gradually decreases, slightly decrease the opening degrees of the air mixing valve, the combustion-supporting air valve, and the gas valve;

[0178] If the actual hot blast stove temperature is close to the target hot blast stove temperature, keep the current opening degree of the air mixing valve unchanged, as well as the opening degrees of the combustion-supporting air valve and the gas valve;

[0179] If the actual hot blast stove temperature is far from the target hot blast stove temperature and the deviation is large, significantly adjust the opening degrees of the air mixing valve, the combustion-supporting air valve, and the gas valve.

[0180] Compare the target hot blast stove temperature identified by the model with the actual hot blast stove temperature. If there is a deviation, the model can be adjusted through the feedback mechanism, improving the accuracy and adaptability.

[0181] Through the above steps, the multi-condition parameter self-tuning model can identify the temperature target values under different conditions in real time, so that the control system of the hot blast stove can adjust the temperature more precisely and adapt to different operating states or environmental conditions.

[0182] The present invention has been described by the above related embodiments. However, the above embodiments are only examples for implementing the present invention. It must be pointed out that the disclosed embodiments do not limit the scope of the present invention. On the contrary, all modifications and refinements made without departing from the spirit and scope of the present invention fall within the scope of patent protection of the present invention.

Claims

1. A temperature intelligent control method for hot blast stoves based on working condition recognition technology and fuzzy PID, characterized in that: It includes the following steps: S1: Collect the historical operation data of the hot blast stove; Collect the historical operation data of the hot blast stove and the temperature of the hot blast stove under different working conditions; The temperature of the hot blast stove includes the actual value of the hot blast stove temperature and the target value of the hot blast stove temperature; The actual value of the hot blast stove temperature includes the actual value of the furnace temperature, the actual value of the temperature behind the furnace, and the actual value of the inlet temperature of the SCR reactor; The target value of the hot blast stove temperature includes the target value of the furnace temperature, the target value of the temperature behind the furnace, and the target value of the inlet temperature of the SCR reactor; The different working conditions refer to different operating states or environmental conditions; S2: Construct a multi-condition matrix; Construct a multi-condition matrix for the temperature control and optimization of the hot blast stove according to the collected historical operation data of the hot blast stove; The rows of the multi-condition matrix correspond to different working conditions, and the columns correspond to each historical operation data; S3: Establish a multi-condition parameter self-tuning model according to the multi-condition matrix; Establish a multi-condition parameter self-tuning model for the temperature control of the hot blast stove according to the multi-condition matrix; S4: Real-time identify the target temperature under the current working condition; A variety of sensors are arranged inside the hot blast stove; Collect the operation data under the current working condition through the sensors in real time, input the operation data under the current working condition into the multi-condition parameter self-tuning model, and the multi-condition parameter self-tuning model identifies the target value of the furnace temperature, the target value of the temperature behind the furnace, and the target value of the inlet temperature of the SCR reactor under the current working condition according to the operation data; S5: Design a control system to regulate the current furnace temperature, temperature behind the furnace, and inlet temperature of the SCR reactor; The control system includes a fuzzy cascade control system for the inlet temperature of the reactor, a fuzzy variable ratio control system for the furnace temperature, and a fuzzy PID control system for the temperature behind the furnace; Control the current furnace temperature, temperature behind the furnace, and inlet temperature of the SCR reactor respectively through the control system combined with the target value of the furnace temperature, the target value of the temperature behind the furnace, and the target value of the inlet temperature of the SCR reactor identified in step S4; Specifically, it includes the following sub-steps: S51: The fuzzy cascade control system for the inlet temperature of the reactor regulates the inlet temperature of the SCR reactor; The main loop of the fuzzy cascade control system for the inlet temperature of the reactor includes a reactor temperature fuzzy controller, a gas flow controller, and a gas valve actuator; S52: The fuzzy PID control system for the furnace temperature regulates the furnace temperature; The fuzzy PID control system for the furnace temperature includes a temperature fuzzy controller, a temperature fuzzy compensation controller, a flow controller, and a combustion-supporting valve actuator; S53: The fuzzy variable ratio control system for the temperature behind the furnace regulates the temperature behind the furnace; The fuzzy variable ratio control system for the temperature behind the furnace includes a temperature controller behind the furnace and an air mixing valve actuator.

2. The intelligent temperature control method of the hot blast stove based on the working condition identification technology and fuzzy PID according to claim 1, characterized in that: In step S3, the establishment of the multi-condition parameter self-tuning model includes the following steps: S31: Feature parameter selection; Select the historical operation data related to the hot blast stove temperature from the historical operation data, and extract the features of the historical operation data; The features include average value, variance, maximum value, and minimum value; S32: Construct an association model; Construct an association model between the basically extracted features and the data of the hot blast stove temperature target value collected in step S1, and denote it as the multi-condition parameter self-tuning model; S33: Perform parameter tuning; Input the historical operation data under different conditions in the multi-condition matrix into the multi-condition parameter self-tuning model, output the predicted hot blast stove temperature target value, compare the difference between the output predicted hot blast stove temperature target value and the actually collected hot blast stove temperature target value, and tune the parameters in the multi-condition parameter self-tuning model through the parameter optimization algorithm.

3. The intelligent temperature control method for hot blast stoves based on condition identification technology and fuzzy PID as claimed in claim 1, characterized in that: Step S51 includes the following sub-steps: S511: Calculate the temperature error and error derivative between the actual value and the target value of the reactor inlet temperature; A temperature sensor is installed inside the reactor inlet, and the real-time actual value of the reactor inlet temperature is collected through the temperature sensor; Obtain the temperature error according to the actual value of the reactor inlet temperature and the reactor inlet temperature target value identified in step S4, and the temperature error is the difference between the reactor inlet temperature target value and the actual value of the furnace temperature; Further, the differential algorithm calculates the error derivative according to the actual values of the reactor inlet temperature at each timestamp in the previous period collected by the temperature sensor and the reactor inlet temperature target values at each timestamp in the previous period identified by the multi-condition parameter self-tuning model; S512: Obtain the gas flow error and error derivative; Input the temperature error and error derivative obtained in step S511 into the reactor temperature fuzzy controller, and the reactor temperature fuzzy controller outputs the gas flow; Further, obtain the gas flow at each timestamp in the previous period through the reactor temperature fuzzy controller; The sensor collects the current actual gas flow in real time; The differential algorithm obtains the gas flow error according to the output gas flow and the current actual gas flow, and obtains the gas flow error derivative according to the actual gas flows at each timestamp in the previous period collected by the sensor and the gas flows at each timestamp in the previous period obtained by the reactor temperature fuzzy controller; S513: Regulate the reactor inlet temperature; Input the gas flow error and gas flow error derivative into the gas flow controller, the gas flow controller adjusts the gas valve opening according to the fuzzy rule, the gas flow controller outputs the adjusted gas valve opening, and input the valve opening into the gas valve actuator to obtain the actual value of the reactor inlet temperature in the next control cycle; Loop steps S511 - S513 to regulate the reactor inlet temperature until the actual value of the reactor inlet temperature is approximately equal to the reactor inlet temperature target value; Further, use the gas pressure as the secondary loop to offset the change in gas pressure by adjusting the gas flow.

4. The intelligent temperature control method for hot blast stoves based on condition identification technology and fuzzy PID as claimed in claim 1, characterized in that: Step S52 includes the following sub-steps: S521: Calculate the temperature error and error derivative between the actual value and the target value of the furnace temperature; A temperature sensor is installed inside the furnace, and the actual value of the real-time furnace temperature is collected through the temperature sensor; A temperature error is obtained based on the actual furnace temperature value and the target furnace temperature value identified in step S4. The temperature error is the difference between the target furnace temperature value and the actual furnace temperature value; Further, the differential algorithm calculates the error derivative based on the actual furnace temperature values at each timestamp in the previous period collected by the temperature sensor and the target furnace temperature values at each timestamp in the previous period identified by the multi-condition parameter self-tuning model; S522: Obtain the current air-fuel ratio setting value; Input the temperature error and the error derivative into the temperature fuzzy controller, and the temperature fuzzy controller outputs the current air-fuel ratio setting value; S523: Output the opening degree of the combustion-supporting air valve in the current control cycle; Calculate the actual value of the combustion-supporting air flow according to the air-fuel ratio and the gas flow rate. The actual value of the combustion-supporting air flow = air-fuel ratio * gas flow rate; The air-fuel ratio is the current air-fuel ratio setting value obtained in step S522, and the gas flow rate is obtained by real-time collection by the sensor inside the furnace; Further, the furnace temperature fuzzy PID control system presets a target value for the combustion-supporting air flow rate. According to the target value of the combustion-supporting air flow rate and the actual value of the combustion-supporting air flow rate, a combustion-supporting air flow rate error is obtained. The combustion-supporting air flow rate error is the difference between the target value of the combustion-supporting air flow rate and the actual value of the combustion-supporting air flow rate; Further, according to the gas flow rate at each timestamp in the previous period collected by the sensor, the actual value of the combustion-supporting air flow rate at each timestamp in the previous period is obtained; Further, the differential algorithm calculates the derivative of the combustion-supporting air flow rate error based on the actual value of the combustion-supporting air flow rate at each timestamp in the previous period and the target value of the combustion-supporting air flow rate preset by the furnace temperature fuzzy PID control system; Input the combustion-supporting air flow rate error and the derivative of the combustion-supporting air flow rate error into the combustion-supporting air flow rate controller, and the combustion-supporting air flow rate controller outputs the opening degree of the combustion-supporting air valve in the current control cycle, denoted as the opening degree u2 of the combustion-supporting air valve; S524: Regulate the furnace temperature; There is also a feed-forward temperature fuzzy compensation controller. Input the temperature error and the error derivative of the furnace temperature obtained in step S521 into the temperature fuzzy compensation controller, adjust the opening degree of the combustion-supporting air valve according to the fuzzy rules, and output the adjusted opening degree of the combustion-supporting air valve, denoted as the opening degree du of the combustion-supporting air valve; Obtain the actual opening degree of the combustion-supporting air valve. The actual opening degree of the combustion-supporting air valve = opening degree u2 of the combustion-supporting air valve + opening degree du of the combustion-supporting air valve; Input the actual opening degree of the combustion-supporting air valve into the combustion-supporting air valve actuator to obtain the actual value of the furnace temperature in the next control cycle; Loop steps S521 - S524 to adjust the furnace temperature until the actual furnace temperature value is approximately equal to the target furnace temperature value.

5. The intelligent temperature control method for the hot blast stove based on the working condition identification technology and fuzzy PID as described in claim 1, characterized in that: Step S53 includes the following sub-steps: S531: Calculate the temperature error and the error derivative between the actual value and the target value of the temperature after the furnace; A temperature sensor is installed inside the hot blast stove, and the actual value of the temperature after the furnace is collected in real time through the temperature sensor; Obtain the temperature error based on the actual value of the post-furnace temperature and the target value of the post-furnace temperature identified in step S4. The temperature error is the difference between the target value of the post-furnace temperature and the actual value of the post-furnace temperature; Further, the differential algorithm calculates the derivative of the post-furnace temperature error based on the actual values of the post-furnace temperature at each time stamp in the previous period collected by the temperature sensor and the target values of the post-furnace temperature at each time stamp identified by the multi-condition parameter self-tuning model; S532: Obtain the opening degree of the air mixing valve; Input the post-furnace temperature error and the derivative of the post-furnace temperature error into the post-furnace temperature controller. The post-furnace temperature controller adjusts the opening degree of the air mixing valve according to the fuzzy rules and outputs the adjusted opening degree of the air mixing valve; Input the output opening degree of the air mixing valve into the air mixing valve actuator, and the air mixing valve actuator outputs the actual value of the post-furnace temperature in the next control cycle; Loop steps S531 - S532 to adjust the furnace temperature until the actual value of the furnace temperature is approximately equal to the target value of the furnace temperature.

6. The intelligent temperature control method for the hot blast stove based on the condition identification technology and fuzzy PID as claimed in claim 3, characterized in that: The fuzzy rules are as follows: If the actual value of the post-furnace temperature is lower than the target value of the post-furnace temperature and the deviation continues to increase, increase the opening degree of the air mixing valve; If the actual value of the post-furnace temperature is lower than the target value of the post-furnace temperature but the deviation gradually decreases, slightly increase the opening degree of the air mixing valve; If the actual value of the post-furnace temperature is higher than the target value of the post-furnace temperature and the deviation continues to increase, decrease the opening degree of the air mixing valve; If the actual value of the post-furnace temperature is higher than the target value of the post-furnace temperature but the deviation gradually decreases, slightly decrease the opening degree of the air mixing valve; If the actual value of the post-furnace temperature is close to the target value of the post-furnace temperature, keep the current opening degree of the air mixing valve unchanged; If the actual value of the post-furnace temperature is far from the target value of the post-furnace temperature and the deviation is large, greatly adjust the opening degree of the air mixing valve.

Images