A desulfurization slurry pool liquid level control method based on intelligent prediction algorithm

Abstract

The application discloses a desulfurization slurry pool liquid level control method based on an intelligent prediction algorithm, wherein the flow of a limestone slurry supplement pump of a wet desulfurization system is obtained by PID calculation based on the deviation of slurry pH and a slurry pH set value, and the feedforward instruction is obtained based on the desulfurization inlet SO2 concentration measurement value, the desulfurization inlet flue gas flow, the slurry circulating pump flow, the slurry drainage pump flow, the slurry pool liquid level and the limestone slurry supplement pump flow; the flow of a demisting flushing pump of the wet desulfurization system is obtained by PID calculation based on the deviation of the slurry pool liquid level and a slurry pool liquid level set value, and the feedforward instruction is obtained based on the slurry circulating pump flow, the slurry drainage pump flow, the limestone slurry supplement pump flow, the unit load and the demisting flushing pump flow; and the precise control of the slurry pool liquid level of the wet desulfurization system is realized by adjusting the running flow of the limestone slurry supplement pump and the demisting flushing pump. The control method can effectively cope with the fluctuation of the desulfurization slurry pool liquid level in the frequent and rapid load change process of the coal-fired unit.

Description

Technical Field

[0001] The present invention relates to the technical field of coal-fired power generation, and in particular to a desulfurization slurry pool liquid level control method based on an intelligent prediction algorithm. Background Art

[0002] In the future, coal-fired power generation units will take on more peak-shaving and frequency-regulating tasks to absorb more wind and solar energy. As a result, coal-fired power generation units will frequently experience large load variations and rapid load variation rates, which will inevitably affect the operation of coal-fired power generation units' desulfurization systems.

[0003] Wet flue gas desulfurization (FGD) systems are the most widely used desulfurization technology in coal-fired power plants, and control of the desulfurization slurry tank is a crucial component. When coal-fired units are subject to frequent load fluctuations, the hysteresis of the desulfurization system makes controlling the slurry tank level and pH more difficult. Current control strategies are unable to cope with these frequent load fluctuations, resulting in poor regulation. Summary of the Invention

[0004] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a desulfurization slurry pool level control method based on an intelligent prediction algorithm to cope with the frequent load changes of coal-fired power generation units, so that the slurry pool level and pH can be better controlled during rapid and frequent load changes.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A desulfurization slurry pool level control method based on intelligent prediction algorithm, flow instruction F of limestone slurry replenishment pump in wet desulfurization system spump-v The deviation between the slurry pH and the slurry pH set value is obtained through PID calculation. The feedforward instruction F2(x) is obtained based on the measured value of SO2 concentration at the desulfurization inlet, the flue gas flow at the desulfurization inlet, the slurry circulation pump flow, the slurry drainage pump flow, the slurry pool level and the limestone slurry replenishment pump flow:

[0007] F2(x)=k1(NNW1(x1,x2,x3,x4,x5,x6)-S ph )

[0008] Where F2(x) is the feedforward of the flow rate instruction of the limestone slurry replenishment pump, m 3 h -1 ; x1 is the measured value of SO2 concentration at the desulfurization inlet, mg m -3 ; x2 is the flue gas flow rate at the desulfurization inlet, kg s -1 ; x3 is the flow rate of the slurry circulation pump, m 3 h -1 ; x4 is the flow rate of slurry drainage pump, m 3 h-1 ; x5 is the slurry pool level, m; x6 is the limestone slurry replenishment pump flow, m 3 h -1 ;S ph is the measured value of slurry pH; NWW1 is intelligent algorithm 1, which predicts slurry pH from x1, x2, x3, x4, x5, and x6 based on the historical operating data of the coal-fired power generation unit; k1 is the adjustment coefficient;

[0009] Flow instruction F of the demisting and flushing pump of the wet flue gas desulfurization system wpump-v The deviation between the slurry pool level and the slurry pool level set value is obtained through PID calculation. The feedforward instruction F1(x) is obtained based on the slurry circulation pump flow, slurry drainage pump flow, limestone slurry replenishment pump flow, unit load and demisting flushing pump flow:

[0010] F1(x)=k2(NNW2(x3,x4,x6,x7,x8)-h slurry )

[0011] Where, F1(x) is the feedforward of the flow command of the demisting flushing pump, m 3 h -1 ; x7 is the load of the coal-fired unit, MW; x8 is the flow rate of the demisting flushing pump, m 3 h -1 ; NWW2 is intelligent algorithm 2, which predicts the slurry pool level based on x3, x4, x6, x7, and x8 according to the historical operating data of the coal-fired unit; k2 is the adjustment coefficient;

[0012] Limestone slurry replenishment pump operating flow F spump And the operating flow rate F of the demisting flushing pump wpump The calculation is as follows:

[0013] F spump =F spump-v +F2(x)

[0014] F wpump =F wpump-v +F1(x)

[0015] By adjusting the limestone slurry supplement pump operating flow F spump And the operating flow rate F of the demisting flushing pump wpump , which can realize precise control of the liquid level in the slurry pool of the wet flue gas desulfurization system.

[0016] Preferably, the intelligent algorithms NWW1 and NWW2 adopt BP neural network algorithm, LSTM neural network algorithm, convolutional neural network algorithm or random forest algorithm.

[0017] Preferably, the limestone slurry replenishment pump is operated at a variable frequency, and its operating flow rate is set to a maximum value and a minimum value;

[0018] v·F≤F≤F

[0019] ss spump s

[0020] Where, F spump is the operating flow of the limestone slurry supplementary pump, m 3 h -1 ; F s is the design flow rate of the slurry replenishment pump, m 3 h -1 ;v s Provides the minimum flow rate change for the slurry replenishment pump.

[0021] Preferably, the demisting and flushing pump is operated at a variable frequency, and its operating flow rate is set to a maximum value and a minimum value;

[0022] v·F≤F≤F

[0023] ww wpump w

[0024] Where, F wpump is the operating flow rate of the demisting flushing pump, m 3 h -1 ; F w is the design flow rate of the demisting flushing pump, m 3 h -1 ;v w The minimum flow change of the demisting flushing pump.

[0025] Preferably, the slurry pH is set to a safe operating range as follows: when the slurry pH exceeds the safe operating range, the value of the adjustment coefficient k1 is increased;

[0026] MS1≤S ph ≤MS2

[0027] Where S ph is the pH value of the wet desulfurization slurry pool; MS1 is the lower limit of the slurry pH safe operating range, generally 4.8 to 5.0; MS2 is the upper limit of the slurry pH safe operating range, generally 5.8 to 6.0.

[0028] Preferably, the safe operating range of the slurry tank liquid level is set as follows: when the liquid level exceeds the safe operating range, the value of the adjustment coefficient k2 is increased;

[0029] MW1≤h slurry ≤MW2

[0030] Where h slurryis the liquid level of the slurry pool in the wet flue gas desulfurization system, m; MW1 is the lower limit of the safe operating range of the slurry pool liquid level, generally taken as 8.0~8.5; MW2 is the upper limit of the safe operating range of the slurry pool liquid level, generally taken as 9.5~10.0.

[0031] Compared with the prior art, the present invention has the following advantages:

[0032] 1) In the process of adjusting the pH value of the desulfurization system slurry, the control method of the present invention introduces the real-time changes of the SO2 concentration measurement value at the desulfurization inlet of the coal-fired unit, the flue gas flow rate at the desulfurization inlet, the slurry circulation pump flow rate, the slurry drainage pump flow rate, the slurry pool liquid level and the limestone slurry replenishment pump flow rate into the feedforward instruction of the desulfurization system slurry pH control through an optimization algorithm. It can intelligently identify the impact of changes in flue gas and slurry parameters on the desulfurization system slurry pH during rapid and frequent load changes of the coal-fired unit.

[0033] 2) The control method of the present invention introduces the real-time changes of the slurry circulation pump flow, the slurry drainage pump flow, the limestone slurry replenishment pump flow, the unit load and the demisting and flushing pump flow into the feedforward instructions of the desulfurization system slurry pool level control during the process of adjusting the desulfurization system slurry pool level, and can intelligently identify the impact of changes in flue gas and slurry parameters of coal-fired units during rapid and frequent load changes on the desulfurization system slurry pool level.

[0034] 3) The control method of the present invention adopts an intelligent prediction algorithm to analyze and predict a large amount of historical data of the wet desulfurization system of the coal-fired unit, taking into account the mutual influence of the slurry pH and the slurry pool liquid level adjustment of the wet desulfurization system of the coal-fired unit. The pH and liquid level values ​​of the desulfurization slurry pool can be obtained in advance, and the limestone slurry replenishment flow and the demisting flushing pump flow are acted on in advance. The adjustment of the slurry pH is taken into account in the process of adjusting the liquid level of the slurry pool of the desulfurization system, thereby achieving better control effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 Schematic diagram of the control method of the present invention. DETAILED DESCRIPTION

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] like Figure 1 As shown, the present invention proposes a desulfurization slurry pool level control method based on intelligent prediction algorithm, the flow instruction F of the limestone slurry replenishment pump of the wet desulfurization system spump-v The deviation Δ1 between the slurry pH and the slurry pH set value is obtained through PID calculation, and its feedforward instruction F2(x) is obtained based on the measured value of SO2 concentration at the desulfurization inlet, the flue gas flow at the desulfurization inlet, the slurry circulation pump flow, the slurry drainage pump flow, the slurry pool level and the limestone slurry replenishment pump flow:

[0038] F2(x)=k1(NNW1(x1,x2,x3,x4,x5,x6)-S ph )

[0039] Where F2(x) is the feedforward of the flow rate instruction of the limestone slurry replenishment pump, m 3 h -1 ; x1 is the measured value of SO2 concentration at the desulfurization inlet, mg m -3 ; x2 is the flue gas flow rate at the desulfurization inlet, kg s -1 ; x3 is the flow rate of the slurry circulation pump, m 3 h -1 ; x4 is the flow rate of slurry drainage pump, m 3 h -1 ; x5 is the slurry pool level, m; x6 is the limestone slurry replenishment pump flow, m 3 h -1 ;S ph is the measured value of slurry pH; NWW1 is intelligent algorithm 1, which predicts slurry pH from x1, x2, x3, x4, x5, and x6 based on the historical operating data of the coal-fired power generation unit; k1 is the adjustment coefficient;

[0040] Flow instruction F of the demisting and flushing pump of the wet flue gas desulfurization system wpump-v The deviation Δ2 between the slurry pool level and the slurry pool level set value is obtained through PID calculation. Its feedforward instruction F1(x) is obtained based on the slurry circulation pump flow, slurry drainage pump flow, unit load, demisting flushing pump flow, and limestone slurry replenishment pump flow:

[0041] F1(x)=k2(NNW2(x3,x4,x6,x7,x8)-h slurry )

[0042] Where, F1(x) is the feedforward of the flow command of the demisting flushing pump, m 3 h -1 ; x7 is the load of the coal-fired unit, MW; x8 is the flow rate of the demisting flushing pump, m 3 h -1 ; NWW2 is intelligent algorithm 2, which predicts the slurry pool level based on x3, x4, x6, x7, and x8 according to the historical operating data of the coal-fired unit; k2 is the adjustment coefficient;

[0043] Limestone slurry replenishment pump operating flow F spump And the operating flow rate F of the demisting flushing pump wpump The calculation is as follows:

[0044] F spump =F spump-v +F2(x)

[0045] Fwpump =F wpump-v +F1(x)

[0046] By adjusting the limestone slurry supplement pump operating flow F spump And the operating flow rate F of the demisting flushing pump wpump , which can realize precise control of the liquid level in the slurry pool of the wet flue gas desulfurization system.

[0047] The intelligent algorithms NWW1 and NWW2 can choose to use BP neural network algorithm, LSTM neural network algorithm, convolutional neural network algorithm and random forest algorithm.

[0048] The slurry replenishment pump is operated with variable frequency, and its operating flow rate is set to maximum and minimum values;

[0049] v·F≤F≤F

[0050] v s ·F s ≤F spump ≤F s

[0051] Where, F spump is the operating flow of the slurry replenishment pump, m 3 h -1 ; F s is the design flow rate of the slurry replenishment pump, m 3 h -1 ;v s Provides the minimum flow rate change for the slurry replenishment pump.

[0052] The demisting and flushing pump is operated by variable frequency, and its operating flow rate is set to maximum and minimum values;

[0053] v·F≤F≤F

[0054] ww wpump w

[0055] Where, F wpump is the operating flow rate of the demisting flushing pump, m 3 h -1 ; F w is the design flow rate of the demisting flushing pump, m 3 h -1 ;v w This is the minimum flow rate change of the demisting and flushing pump. This value can ensure that the demisting and flushing pump can adjust the operating flow rate under the premise of safe operation.

[0056] The safe operating range of slurry pH is set as follows. When the slurry pH exceeds the safe operating range, increase the value of the adjustment coefficient k1;

[0057] MS1≤S ph ≤MS2

[0058] Where S ph It is the pH value of the wet flue gas desulfurization slurry pool; MS1 is the lower limit of the safe operating range of the slurry pH, which is generally 4.8-5.0. If the slurry pH is too low, it will be detrimental to the removal efficiency of the wet flue gas desulfurization system and cause the slurry circulation pump to consume a lot of power; MS2 is the upper limit of the safe operating range of the slurry pH, which is generally 5.8-6.0. If the slurry pH is too high, the slurry will easily scale, causing problems such as blockage of pumps and pipelines.

[0059] Liquid level height of slurry pool in wet flue gas desulfurization system h slurry Set the safe operating range as follows. When the liquid level exceeds the safe operating range, increase the value of the adjustment coefficient k2;

[0060] MW1≤h slurry ≤MW2

[0061] Where h slurry is the liquid level of the slurry pool in the wet flue gas desulfurization system, m; MW1 is the lower limit of the safe operating range of the slurry pool liquid level, which is generally 8.0-8.5. Too low a slurry pool liquid level is not conducive to the safe operation of the slurry circulation pump; MW2 is the upper limit of the safe operating range of the slurry pool liquid level, which is generally 9.5-10.0. The height of the slurry pool liquid level must be lower than the design height of the slurry pool, and a certain safety margin must be retained.

Claims

1. A desulfurization slurry pool level control method based on an intelligent prediction algorithm, characterized by: Flow instruction F of limestone slurry replenishment pump in wet flue gas desulfurization system spump-v The deviation between the slurry pH and the slurry pH set value is obtained through PID calculation. The feedforward instruction F2(x) is obtained based on the measured value of SO2 concentration at the desulfurization inlet, the flue gas flow at the desulfurization inlet, the slurry circulation pump flow, the slurry drainage pump flow, the slurry pool level and the limestone slurry replenishment pump flow: F2(x)=k1(NNW1(x1,x2,x3,x4,x5,x6)-S ph ) Where F2(x) is the feedforward of the flow rate instruction of the limestone slurry replenishment pump, m 3 h -1 ; x1 is the measured value of SO2 concentration at the desulfurization inlet, mg m -3 ; x2 is the flue gas flow rate at the desulfurization inlet, kg s -1 ; x3 is the flow rate of slurry circulation pump, m 3 h -1 ; x4 is the flow rate of slurry drainage pump, m 3 h -1 ; x5 is the slurry tank level, m; x6 is the flow rate of limestone slurry supplement pump, m 3 h -1 ; S ph is the measured value of slurry pH; NWW1 is intelligent algorithm 1, which predicts slurry pH from x1, x2, x3, x4, x5, and x6 based on the historical operating data of the coal-fired power generation unit; k1 is the adjustment coefficient; Flow instruction F of the demisting and flushing pump of the wet flue gas desulfurization system wpump-v The slurry pool level h slurry The deviation from the slurry tank level set point is obtained through PID calculation, and its feedforward instruction F1(x) is obtained based on the slurry circulation pump flow, slurry drainage pump flow, limestone slurry replenishment pump flow, unit load and demisting flushing pump flow: F1(x)=k2(NNW2(x3,x4,x6,x7,x8)-h slurry ) Where, F1(x) is the feedforward of the flow command of the demisting flushing pump, m 3 h -1 ;h slurry is the liquid level of the slurry pool of the wet flue gas desulfurization system, m; x7 is the load of the coal-fired unit, MW; x8 is the flow rate of the demisting flushing pump, m 3 h -1 ; NWW2 is intelligent algorithm 2, which predicts the slurry pool level based on x3, x4, x6, x7, and x8 according to the historical operating data of the coal-fired unit; k2 is the adjustment coefficient; Limestone slurry replenishment pump operating flow F spump And the operating flow rate F of the demisting flushing pump wpump The calculation is as follows: F spump =F spump-v +F2(x) F wpump =F wpump-v +F1(x) By adjusting the limestone slurry supplement pump operating flow F spump And the operating flow rate F of the demisting flushing pump wpump , to achieve precise control of the liquid level in the slurry pool of the wet flue gas desulfurization system.

2. The desulfurization slurry pool level control method based on intelligent prediction algorithm according to claim 1 is characterized in that: The intelligent algorithms NWW1 and NWW2 use BP neural network algorithm, LSTM neural network algorithm, convolutional neural network algorithm or random forest algorithm.

3. The desulfurization slurry pool level control method based on intelligent prediction algorithm according to claim 1 is characterized in that: The limestone slurry replenishment pump is variable frequency operated, and its operating flow rate is set to maximum and minimum values; v s ·F s ≤F spump ≤F s Where, F spump is the operating flow of the limestone slurry supplementary pump, m 3 h -1 ; F s is the design flow rate of the slurry replenishment pump, m 3 h -1 ;v s Provides the minimum flow rate change for the slurry replenishment pump.

4. The desulfurization slurry pool level control method based on intelligent prediction algorithm according to claim 1 is characterized in that: The demisting and flushing pump is operated by variable frequency, and its operating flow rate is set to maximum and minimum values; v w ·F w ≤F wpump ≤F w Where, F wpump is the operating flow rate of the demisting flushing pump, m 3 h -1 ; F w is the design flow rate of the demisting flushing pump, m 3 h -1 ;v w The minimum flow change of the demisting flushing pump.

5. The desulfurization slurry pool level control method based on intelligent prediction algorithm according to claim 1 is characterized in that: The safe operating range of slurry pH is set as follows. When the slurry pH exceeds the safe operating range, increase the value of the adjustment coefficient k1; MS1≤S ph ≤MS2 Wherein, MS1 is the lower limit of the safe operating range of slurry pH, which is 4.8 to 5.0; MS2 is the upper limit of the safe operating range of slurry pH, which is 5.8 to 6.

0.

6. The desulfurization slurry pool level control method based on intelligent prediction algorithm according to claim 1 is characterized in that: The safe operating range of the liquid level of the slurry pool of the wet flue gas desulfurization system is as follows. When the liquid level exceeds the safe operating range, increase the value of the adjustment coefficient k2; MW1≤h slurry ≤MW2 In the formula, MW1 is the lower limit of the safe operating range of the slurry pool liquid level height, which is 8.0~8.5; MW2 is the upper limit of the safe operating range of the slurry pool liquid level height, which is 9.5~10.0.

Images