A method for coordinated control of a coal gas generator set and an entire coal gas pipeline network

Abstract

The application discloses a coal gas generator set and whole coal gas pipe network coordinated control method, a basic digital twin model of the coal gas pipe network is established on a computer, parameters are introduced to the basic model, a multi-parameter BP neural network is used for correction, the preparation of the correction is judged through the error between the measured value and the predicted value, and a high-precision digital twin model is obtained through long-period training; total coal gas consumption of the coal gas generator set is measured; the total coal gas consumption is smoothed and optimized to obtain smoothed total coal gas consumption of the coal gas generator set; the smoothed and optimized total coal gas consumption is taken as a benchmark to construct load optimization distribution of multiple coal gas generator sets; the coal gas consumption distributed by the multiple sets is taken as the benchmark, and the pressure of the coal gas pipe network is taken as a stable target to realize the establishment of the unit coordinated control method, and the self-adaptive coordinated control of the coal gas generator set is constructed with the controllable pressure of the coal gas pipe network as a target, so that the large fluctuation of the pressure of the coal gas pipe network is avoided, and the comprehensive benefit of the coal gas generator set is improved.

Description

Technical Field

[0001] This invention relates to a coordinated control method, and more specifically, to a coordinated control method for a gas generator set and the entire gas pipeline network, belonging to the field of power generation. Background Technology

[0002] In recent years, steel smelting enterprises have invested in and constructed gas-fired power plants. The gas is mainly produced by blast furnaces in the smelting process, and the entire gas pipeline network connects multiple gas users. The gas consumption of the power plant is directly related to the smelting process and the consumption of each user. Because the gas boiler needs to be coordinated with the overall steelmaking process, predicting the gas consumption of the power plant is difficult, and implementing automation is also challenging. Currently, blast furnace gas power plants generally suffer from poor basic automation. There is no coordination between boiler combustion adjustments and changes in the gas pipeline network, and no coordination between the boiler and the turbine. Adjustments are mainly performed by rough-and-ready operators. Therefore, the following problems arise during production: 1) Large fluctuations or even venting of gas pipeline pressure; 2) Delayed power plant adjustments and low boiler operating efficiency. Summary of the Invention

[0003] The technical problem to be solved by this invention is to overcome the problem that changes in the amount of gas that the gas generator set needs to consume due to changes in the smelting process lead to large fluctuations in the gas pipeline network pressure, which ultimately results in poor economic performance of the generator set. This invention provides a method for coordinated control of the gas generator set and the entire gas pipeline network.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0005] A method for coordinated control of a gas generator set and the entire gas pipeline network, the method comprising the following steps:

[0006] Step 1: Establish a basic digital twin model of the gas pipeline network on the computer, and introduce actual operating parameters to calibrate the basic model;

[0007] Step 2: The digital twin model is calibrated using a multi-parameter BP neural network. The accuracy of the calibration is judged by the error between the measured and predicted values. A high-precision digital twin model is obtained through long-term training.

[0008] Step 3: Predict the total gas consumption of the gas generator set using a high-precision digital twin model;

[0009] Step 4: Use filtering to smooth and optimize the total gas consumption to obtain a smooth total gas consumption of the gas generator set. Establish a basic model of the efficiency parameters and benefit parameters of multiple gas generator sets at full load on the computer.

[0010] Step 5: Using the smoothed and optimized total gas consumption as a benchmark, construct a load optimization allocation for multiple gas generator units; using the gas consumption allocated to multiple units as a benchmark and the gas pipeline pressure as a stability target, establish a unit coordinated control method.

[0011] Preferably, the correction algorithm includes the following steps:

[0012] Step 1: Initialize the weight threshold;

[0013] Step 2: Input training samples and calculate the input and output of each neuron in the hidden layer;

[0014] Step 3: Calculate the input and output of the output layer neurons;

[0015] Step 4: Calculate the output layer error;

[0016] Step 5: Calculate the hidden layer error;

[0017] Step 6: Adjust the weights and thresholds of the output layer and hidden layers;

[0018] Step 7: Confirm that all sample data training is complete. If training is complete, check whether the error is less than the predetermined value. If the error is less than the predetermined value, the current round of training ends. If the error is not less than the predetermined value, update the total number of training steps and re-enter the initial weight threshold to repeat until training ends. If training is not complete, re-enter the initial weight threshold to repeat until training ends.

[0019] Preferably, the target value of gas consumption of multiple units is updated every 60 seconds.

[0020] Preferably, the gas pipeline network includes blast furnace No. 1, blast furnace No. 2, hot blast stove of blast furnace No. 1, hot blast stove of blast furnace No. 2, pulverized coal injection No. 1, pulverized coal injection No. 2, bar heating furnace, wire rod heating furnace, sintering, and excess venting pipe; the gas generator set includes an 80MW power plant and a 93MW power plant.

[0021] Blast furnaces No. 1 and No. 2 are both connected to a main pipeline. The main pipeline is divided into four pipes: Pipeline 1, Pipeline 2, Pipeline 3, and Pipeline 4. Pipelines 1 and 3 are connected to the hot blast stoves of blast furnaces No. 1 and No. 2, respectively. Pipeline 2 is connected to pulverized coal injection stations No. 1 and No. 2. At the end of Pipeline 2, there are sub-pipelines for connecting to a 40MW power plant, a pelletizing plant, and a denitrification gas desulfurization plant. Pipeline 2 also has a sub-pipeline for sintering ladles.

[0022] The No. 4 pipeline is divided into multiple sub-pipelines, which are respectively connected to the 80MW power plant, the 93MW power plant, the excess venting pipe, the bar heating furnace, the wire heating furnace, and the sintering plant.

[0023] Preferably, it also includes an automatic combustion method for the gas generator set: establishing a functional model of gas volume and air volume, calculating the air volume command based on the real-time gas volume, and achieving closed-loop regulation of the blower by controlling the air volume command and the actual air volume.

[0024] A cross-limiting function for air supply volume and gas volume is established. The gas volume regulating valve opening command is compared with the cross-limiting function by taking the smaller value, and the final gas volume regulating valve command is output to control the opening of the gas volume regulating valve on site.

[0025] Preferably, the gas generator set further includes a variable frequency blower, which uses the gas volume as feedforward for the variable frequency control of the induced draft fan to establish a control loop between the variable frequency of the induced draft fan and the negative pressure of the furnace.

[0026] The forced draft fan frequency conversion command serves as feedforward for the induced draft fan frequency conversion control, achieving closed-loop control of the induced draft fan by using the furnace negative pressure as the control target. A linear function of gas volume and forced draft volume is established, and forced draft volume commands are output in real time to control the air volume of the forced draft fan.

[0027] Preferably, a function is established for the gas quantity and oxygen quantity commands. The oxygen quantity command and the actual oxygen quantity are corrected by controlling the air supply quantity command coefficient, and the air supply quantity command coefficient is adjusted in the range of 0.8-1.2.

[0028] Preferably, it also includes an automatic combustion method for gas generator sets: taking the steam drum water level as the control object, establishing single-impulse and three-impulse control strategies for the feedwater pump, wherein the three impulses are the steam drum water level, feedwater flow rate, and main steam flow rate, and the single-impulse and three-impulse control modes are automatically switched according to the load conditions.

[0029] The feedwater pump is a variable frequency feedwater pump: with the steam drum water level as the control target, a three-impulse control loop is established between the feedwater pump frequency conversion, the steam drum water level, the main feedwater flow rate, and the main steam flow rate.

[0030] Preferably, the gas generator set includes a steam turbine generator set, and the coordinated automatic method for the steam turbine generator set is as follows: a function is established between the amount of gas fed into the furnace and the target value of the sliding pressure of the steam turbine unit; the target value of the sliding pressure of the unit is calculated by the real-time amount of gas fed into the furnace; and the main steam pressure is closed-loop controlled by the steam turbine DEH to control the load of the steam turbine generator set.

[0031] Beneficial effects: It can predict and calculate the total target value of coal gas consumption in power plant boilers in real time, and simultaneously optimize the total target value of coal gas consumption in power plant boilers. It establishes an optimal allocation model for coal gas consumption of multiple units with comprehensive benefits as the goal. Based on coal gas consumption and with the goal of controllable coal gas pipeline pressure, it constructs an adaptive coordinated control of coal gas generator units to avoid large fluctuations in coal gas pipeline pressure and improve the comprehensive benefits of coal gas generator units. Attached Figure Description

[0032] Figure 1 A diagram of the gas pipeline network system in Embodiment 1 of the present invention.

[0033] Figure 2 A flowchart of the coordinated control method between the gas generator set and the entire gas pipeline network in Embodiment 1 of the present invention.

[0034] Figure 3 Flowchart of the digital twin BP correction algorithm for gas pipeline network in Embodiment 1 of the present invention.

[0035] Figure 4 The unit load target value smoothing optimization curve of Embodiment 1 of the present invention.

[0036] Figure 5 The power plant gas consumption instruction generation logic of Embodiment 1 of the present invention. Detailed Implementation

[0037] The present invention will be further described below with reference to the accompanying drawings, but the present invention is not limited to the following embodiments.

[0038] Because the gas generator set lacks an automatic control scheme coordinated with the entire gas pipeline network, it cannot correctly, promptly, and effectively adjust to pressure fluctuations in the gas pipeline network. When the amount of gas produced in the smelting process and the amount of other gas consumed in the pipeline network change significantly, problems such as large pressure fluctuations in the gas pipeline network, even venting, or low pressure that can affect production are likely to occur. Furthermore, large pressure fluctuations in the gas pipeline network lead to large variations in the amount of gas fed into the furnace, while adjustments to the air supply, water supply, and furnace negative pressure lag behind, which is detrimental to the safe, stable, and efficient operation of the gas generator set.

[0039] Example 1

[0040] like Figure 1 As shown, the gas pipeline network in this embodiment includes blast furnace No. 1, blast furnace No. 2, hot blast stove of blast furnace No. 1, hot blast stove of blast furnace No. 2, pulverized coal injection No. 1, pulverized coal injection No. 2, bar heating furnace, wire rod heating furnace, sintering, and excess venting pipe; the gas generator set includes an 80MW power plant and a 93MW power plant.

[0041] like Figure 2 As shown, the coordinated control method of the gas generator set and the entire gas pipeline network in this embodiment realizes the prediction of the total target value of gas consumption of the gas power plant in the gas pipeline network, the smoothing optimization of the total target value of gas consumption, the allocation of gas consumption of the two gas generator sets through the multi-furnace multi-machine load optimization allocation model, and the construction of coordinated control of the gas generator set based on the allocated gas consumption.

[0042] The control method specifically includes the following steps:

[0043] Step 100: Establish a basic digital twin model of the entire gas pipeline network on a computer, and introduce actual operating parameters to calibrate the basic model. The digital twin model uses a multi-parameter BP neural network for calibration, and the calibration algorithm flowchart is shown below. Figure 3 As shown, the readiness of the correction is judged by the error between the measured and predicted values, and a high-precision digital twin model is obtained through long-term training. The total gas consumption of the gas generator unit is predicted using the high-precision digital twin model (mechanism model and data model).

[0044] Step 110: Filtering is performed using IFFT(FFT(x).*FFT(F)) to obtain a smoothed total gas consumption of the gas generator unit. Due to fluctuations in gas pipeline parameters, the total gas consumption signal of the gas generator unit obtained from the digital twin model will have some oscillations; therefore, filtering is performed. Figure 4 The figure shows the smoothing optimization curve of the unit load target value. A basic model of the efficiency and benefit parameters of the two gas generator units under full load was established on a computer.

[0045] Step 120: Using the smoothed optimized total gas consumption target of the gas generator sets as a benchmark, construct an optimized load allocation system for multiple gas generator sets. Specifically: input the smoothed total gas consumption of the gas generator sets into the unit load optimization allocation system to obtain the benchmark values ​​of gas consumption for two units. Considering system stability, in this embodiment, the target value of unit gas consumption is updated every 60 seconds; Figure 5 As shown, the optimized allocation logic for the unit's gas quantity command includes a prediction system and a pre-control system. The prediction system is used to predict the total gas quantity (predicting the total gas consumption of the gas generator unit through a high-precision digital twin model) and calculate the gas consumption of each part of the gas pipeline network (using filtering to smooth and optimize the total gas consumption to obtain a smooth total gas consumption of the gas generator unit). The data is then allocated and calculated for distribution to the 80MW and 93MW power plants. The pre-control system sets the gas quantity benchmarks for the 80MW and 93MW power plants respectively, and implements error (PID) change feedback through the gas pipeline network pressure and the gas pipeline network pressure setting. Finally, the system uses a microcontroller (MUL) to issue gas quantity commands for the 80MW and 93MW power plants, thereby controlling the gas quantity changes and adjustments of the 80MW and 93MW power plants.

[0046] Step 130: Based on the target value of gas consumption of each unit, and with the gas pipeline pressure as the control target, realize the automatic combustion of gas generator units.

[0047] In this embodiment, the target value of the gas generator set's automatic fuel control reference quantity is converted into the gas volume adjustment valve opening command reference through a function. Since the gas volume adjustment valve used on site is a butterfly valve with poor adjustment characteristics, a fuzzy table of the gas volume adjustment valve correction quantity is established by using the gas pipeline pressure and pressure change rate. Through debugging, the coordinated adjustment of the gas volume adjustment valve and the gas pipeline network is achieved.

[0048] In this embodiment, the automatic combustion method for the gas generator set further includes:

[0049] A functional model of gas volume and air volume is established. The air volume command is calculated based on the real-time gas volume. The air volume command and the actual air volume are controlled to achieve closed-loop regulation of the blower.

[0050] In order to make the air supply volume more reasonable, this embodiment establishes a function of gas volume and oxygen volume command. The oxygen volume command and the actual oxygen volume are corrected by controlling the air supply volume command coefficient. The air supply volume command coefficient adjustment range is 0.8-1.2.

[0051] In this embodiment, the automatic combustion method for the gas generator set further includes:

[0052] The frequency conversion command of the forced draft fan serves as the feedforward for the frequency conversion control of the induced draft fan. By using the furnace negative pressure as the control target, the closed-loop control of the induced draft fan is achieved.

[0053] In this embodiment, the automatic combustion method for the gas generator set further includes:

[0054] Using the steam drum water level as the control object, a single-impulse and three-impulse control strategy for the feedwater pump is established. The three impulses are the steam drum water level, feedwater flow rate, and main steam flow rate. The single-impulse and three-impulse control modes are automatically switched according to the load conditions.

[0055] This embodiment also includes automatic coordination of the steam turbine generator set, establishing a function relating the furnace gas flow rate to the unit's sliding pressure target value. The unit's sliding pressure target value is calculated using the real-time furnace gas flow rate, and closed-loop control of the main steam pressure is achieved through the turbine's DEH (Deep Energy Heating) system. For example, based on the pressure deviation value of the gas pipeline network (the difference between the actual value and the set value), the required change in gas flow rate is calculated using the following formula:

[0056]

[0057] Where: ΔV is the change in gas volume, Nm3;

[0058] ΔP is the gas pressure deviation value, in kPa;

[0059] V represents the total storage capacity of the gas pipeline network, in m³.

[0060] R is the molar gas constant, typically taken as 8.314.

[0061] T is the gas temperature, in K;

[0062] 22.4 represents the molar volume of gas under standard conditions, in L / Nm³.

[0063] Finally, it should be noted that the present invention is not limited to the above embodiments, and many variations are possible. All variations that can be directly derived or conceived by those skilled in the art from the disclosure of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for coordinated control of a gas generator set and the entire gas pipeline network, characterized in that the coordinated control method includes the following steps: Step 1: Establish a basic digital twin model of the gas pipeline network on the computer, and introduce actual operating parameters to calibrate the basic model; Step 2: The digital twin model is calibrated using a multi-parameter BP neural network. The accuracy of the calibration is judged by the error between the measured and predicted values. A high-precision digital twin model is obtained through long-term training. Step 3: Predict the total gas consumption of the gas generator set using a high-precision digital twin model; Step 4: Use filtering to smooth and optimize the total gas consumption to obtain a smooth total gas consumption of the gas generator set. Establish a basic model of the efficiency and benefit parameters of multiple gas generator sets at full load on the computer. Step 5: Using the smoothed and optimized total gas consumption as a benchmark, construct a load optimization allocation for multiple gas generator units; using the gas consumption allocated to multiple units as a benchmark and the gas pipeline network pressure stability as the goal, establish a unit coordination control method. The correction algorithm includes the following steps: Step 1: Initialize the weight threshold; Step 2: Input training samples and calculate the input and output of each neuron in the hidden layer; Step 3: Calculate the input and output of the output layer neurons; Step 4: Calculate the output layer error; Step 5: Calculate the hidden layer error; Step 6: Adjust the weights and thresholds of the output layer and hidden layers; Step 7: Confirm that all sample data training is complete. If training is complete, check if the error is less than the predetermined value. If the error is less than the predetermined value, the current training round ends. If the error is not less than the predetermined value, update the total number of training steps and restart the initialization of weights and thresholds until training ends. If training is not complete, restart the initialization of weights and thresholds until training ends. The target values ​​for gas consumption of multiple generating units are updated every 60 seconds; The gas pipeline network includes Blast Furnace No. 1, Blast Furnace No. 2, Hot Blast Stove of Blast Furnace No. 1, Hot Blast Stove of Blast Furnace No. 2, Pulverized Coal Injection System No. 1, Pulverized Coal Injection System No. 2, Bar Heating Furnace, Wire Rod Heating Furnace, Sintering and Excess Venting Pipe; the gas generator sets include an 80MW power plant and a 93MW power plant. Blast furnaces No. 1 and No. 2 are both connected to a main pipeline. The main pipeline is divided into four pipes: Pipeline 1, Pipeline 2, Pipeline 3, and Pipeline 4. Pipelines 1 and 3 are connected to the hot blast stoves of blast furnaces No. 1 and No. 2, respectively. Pipeline 2 is connected to pulverized coal injection stations No. 1 and No.

2. At the end of Pipeline 2, there are sub-pipelines for connecting to a 40MW power plant, for connecting to pellets, and for connecting to denitrification gas. Pipeline 2 also has a sub-pipeline for sintering ladles. The No. 4 pipeline is divided into multiple sub-pipelines, which are respectively connected to the 80MW power plant, the 93MW power plant, the excess venting pipe, the bar heating furnace, the wire heating furnace, and the sintering plant.

2. The method for coordinated control of a gas generator set and the entire gas pipeline network according to claim 1 is characterized in that it further includes an automatic combustion method for the gas generator set: establishing a functional model of gas volume and air supply volume, calculating the air supply volume command through real-time gas volume, and realizing closed-loop regulation of the blower by controlling the air supply volume command and the actual air supply volume.

3. The method for coordinated control of a gas generator set and the entire gas pipeline network according to claim 2 is characterized in that the gas generator set further includes a variable frequency blower, and the gas volume is used as feedforward for the variable frequency control of the induced draft fan to establish a control loop between the variable frequency of the induced draft fan and the negative pressure of the furnace.

4. A method for coordinated control of a gas generator set and the entire gas pipeline network according to claim 2 or 3, characterized in that a function of gas quantity and oxygen quantity command is established, and the air supply quantity command coefficient is corrected by controlling the oxygen quantity command and the actual oxygen quantity, and the air supply quantity command coefficient adjustment range is 0.8-1.

2.

5. The method for coordinated control of a gas generator set and the entire gas pipeline network according to claim 1, characterized in that it further includes an automatic combustion method for the gas generator set: taking the steam drum water level as the control object, establishing a single-impulse and three-impulse control strategy for the feedwater pump, wherein the three impulses are the steam drum water level, the feedwater flow rate and the main steam flow rate, and the single-impulse and three-impulse control modes are automatically switched according to the load conditions.

6. The method for coordinated control of a gas generator set and the entire gas pipeline network according to claim 1, characterized in that the gas generator set includes a steam turbine generator set, and the automatic coordination method for the steam turbine generator set is as follows: a function is established between the amount of gas fed into the furnace and the target value of the sliding pressure of the steam turbine unit; the target value of the sliding pressure of the unit is calculated by the real-time amount of gas fed into the furnace; and the main steam pressure closed-loop control is achieved through the steam turbine DEH to control the load of the steam turbine generator set.

Images