Optimal control method for ID fan of aluminum hydroxide roasting furnace based on multi-model linear interpolation

Abstract

The invention discloses an aluminum hydroxide roasting furnace ID fan optimization control method based on multi-model linear interpolation, and the method comprises the steps: setting three working modes in a process control system: a low load, a medium load and a high load, and the corresponding models are G1, G2 and G3 respectively; the current aluminum hydroxide material discharging amount is input, the system automatically assigns the corresponding ID draught fan rotating speed as a dispatching variable wi, and when w is smaller than or equal to 50% of the ID draught fan rated rotating speed, the low-load model G1 is completely used; when w is equal to 70% ID of the rated rotating speed of the fan, the medium load model G2 is completely used; when w is larger than or equal to 90% ID of the rated rotating speed of the draught fan, the high-load model G3 is completely used, and when the dispatching variable w of the rated rotating speed of the draught fan ranges from 50% to 70%, the process control system uses the 33.33% low-load model and the 66.67% medium-load model. Not only can the aluminum hydroxide produced by the roasting furnace be kept in a suspended state, but also the excess air proportion can be effectively controlled within the range of 1.0-1.5 specified by the operation specification, and the optimal control effect is achieved.

Description

Technical Field

[0001] The invention relates to a fan optimization control method, in particular to a multi-model linear interpolation ID fan optimization control method for an aluminum hydroxide roasting furnace. Background Art

[0002] The aluminum hydroxide roasting furnace process is a production system with complex process characteristics. There are many parameter factors to be controlled during production, such as natural gas flow, ID fan speed, aluminum hydroxide feeding amount, ash return amount (circulation amount), oxygen content, etc. The various parameters influence and relate to each other, and have the characteristics of nonlinearity, hysteresis, and large operating disturbance. The control in the production process mainly relies on manual operation, which is subject to control lag, subjectivity, blindness, and inaccurate control, which often leads to process parameters not reaching the expected targets. For example, the temperature fluctuation of the main furnace can reach more than ±50℃, the oxygen content of the exhaust gas is 3.0~5.7, and the excess air system shows that it exceeds the standard, which is far greater than the 1.0~1.5 specified in the production operation procedures, which directly affects the energy saving and consumption reduction of the roasting furnace and the economic benefits.

[0003] During the production process of aluminum hydroxide roasting furnace, the material aluminum hydroxide needs to be kept in a suspended state. The power source of the material suspension is the ID fan suction. At the same time, the air sucked in by the ID fan is used as combustion air to fully burn with natural gas. Therefore, it is necessary to maintain a certain excess air ratio, such as 1.0 to 1.5 as specified in general operating procedures. That is, the controlled model of the ID fan speed of the aluminum hydroxide roasting furnace often changes with the load. When the load is low, the roasting furnace has low combustion efficiency and slow response; when the load reaches near the rated value, the roasting furnace has high combustion efficiency and fast response. Summary of the invention

[0004] The purpose of the present invention is to provide a multi-model linear interpolation aluminum hydroxide roasting furnace ID fan optimization control method. The method of the present invention can not only keep aluminum hydroxide in a suspended state during the production process of the roasting furnace, but also effectively control the excess air ratio within the range of 1.0 to 1.5 specified in the operating regulations, thereby achieving the best control effect.

[0005] The technical solution of the present invention is as follows: an optimization control method for ID fans of aluminum hydroxide roasting furnaces by multi-model linear interpolation, wherein three working modes are set in the process control system, namely, low load, medium load, and high load. When the load is low, the actual speed of the ID fan w1 is 50% of the rated speed of the ID fan, and the corresponding model is G1; when the load is medium, the actual speed of the ID fan w2 is 70% of the rated speed of the ID fan, and the corresponding model is G2; when the load is high, the actual speed of the ID fan w3 is 90% of the rated speed of the ID fan, and the corresponding model is G3;

[0006] During production, the production operator inputs the current amount of aluminum hydroxide material discharged, and the process control system automatically specifies the ID fan speed corresponding to the current amount of aluminum hydroxide discharged as the scheduling variable w i When the scheduling variable w≤50%ID fan rated speed, the process control system will completely use the low load model G1; when the scheduling variable w=70%ID fan rated speed, the process control system will completely use the medium load model G2; when the scheduling variable w≥90%ID fan rated speed, the process control system will completely use the high load model G3. When the fan rated speed scheduling variable w is between 50% and 70%, the process control system uses 33.33% of the low load model, 66.67% of the medium load model and 0.0% of the high load model.

[0007] In the aforementioned multi-model linear interpolation aluminum hydroxide roasting furnace ID fan optimization control method, the process control system obtains the current control model by weighting through the following formula:

[0008] G0=α1(w)G1+α2(w)G2+α3(w)G3 (1)

[0009]

[0010]

[0011] Beneficial effects of the present invention: Compared with the prior art, the present invention provides a method that allows production operators to splice models between working points by linear interpolation, i.e., multiple working points where the actual speed of the ID fan changes according to scheduling instructions, i.e., the amount of aluminum hydroxide material discharged. In addition, multiple groups of local models of the ID fan draft corresponding to the working points are used to change the actual speed of the ID fan. During the production process, the production operator only needs to input the current amount of aluminum hydroxide material discharged, and the process control system automatically matches and obtains the corresponding actual speed of the ID fan. The current corresponding automatic control model can be confirmed by the actual speed of the ID fan. This multi-model linear interpolation ID fan optimization control method for aluminum hydroxide roasting furnaces can not only keep the aluminum hydroxide produced by the roasting furnace in a suspended state, but also effectively control the excess air ratio within the range of 1.0 to 1.5 specified in the operating procedures, thereby achieving the best control effect. DETAILED DESCRIPTION

[0012] The present invention will be further described below in conjunction with the embodiments, but they are not intended to limit the present invention.

[0013] In the embodiment of the present invention, a multi-model linear interpolation ID fan optimization control method for aluminum hydroxide roasting furnace is used. The production operator selects three working points, namely, low load, medium load, and high load, according to the scheduling instruction, that is, the current aluminum hydroxide feeding amount of the roasting furnace and the corresponding ID fan speed as the scheduling variable. As shown in the following table.

[0014] name <![CDATA[Operating point ID, actual rotational speed w of the fan i > <![CDATA[Local model G i > Remark Low load <![CDATA[w1 = 50% of the rated speed of the ID fan]]> <![CDATA[G1]]> Medium load <![CDATA[w2 = 70% of the rated speed of the ID fan]]> <![CDATA[G2]]> High load <![CDATA[w3 = 90% of the rated speed of the ID fan]]> <![CDATA[G3]]>

[0015] The production operator only needs to input the current amount of aluminum hydroxide material into the system through the human-machine interface of the control software according to the scheduling instruction, and the control strategy automatically specifies the ID fan speed load as the scheduling variable w i . The production process is continuous, and the amount of aluminum hydroxide material discharged is usually variable due to factors such as the upper and lower processes. The corresponding ID fan speed load scheduling variable is usually continuously changing. When the scheduling variable w≤50%ID fan rated speed, the process control system will completely use the low load model G1; when the scheduling variable w=70%ID fan rated speed, the process control system will completely use the medium load model G2; when the scheduling variable w≥90%ID fan rated speed, the process control system will completely use the high load model G3. When the fan rated speed scheduling variable is between 50% and 70%, such as 63.33%, the process control system uses a low load model of 33.33%, a medium load model of 66.67% and a high load model of 0.0%.

[0016] The process control system obtains the current model by weighting through the following formula:

[0017] G0=α1(w)G1+α2(w)G2+α3(w)G3 (1)

[0018]

[0019]

[0020] The process control system performs optimal control based on the weighted model G0.

[0021] Through this multi-model linear interpolation aluminum hydroxide roasting furnace ID fan optimization control method, the aluminum hydroxide produced by the roasting furnace can be kept in a suspended state, and the excess air ratio can be effectively controlled within the range of 1.0 to 1.5 specified in the operating procedures, thereby achieving the best control effect.

[0022] The multi-model linear interpolation optimization control method for the ID fan of the aluminum hydroxide roasting furnace designed by the present invention only describes the ID fan at low load, medium load and high load. If the designer specifies more working points and working point models, the calculation method is similar. The present invention does not require the addition of any instrument equipment and can be easily implemented when configured in the control system.

Claims

1. A multi-model linear interpolation ID fan optimization control method for aluminum hydroxide roasting furnace, characterized by: Three working modes are set in the process control system, namely low load, medium load and high load. When the load is low, the actual speed of the ID fan w1 is 50% of the rated speed of the ID fan, and the corresponding model is G1; when the load is medium, the actual speed of the ID fan w2 is 70% of the rated speed of the ID fan, and the corresponding model is G2; when the load is high, the actual speed of the ID fan w3 is 90% of the rated speed of the ID fan, and the corresponding model is G3; During production, the production operator inputs the current amount of aluminum hydroxide material discharged, and the process control system automatically specifies the ID fan speed corresponding to the current amount of aluminum hydroxide discharged as the scheduling variable w i , when the dispatch variable w≤50%ID fan rated speed, the process control system will fully use the low-load model G1; When the dispatch variable w = 70% ID fan rated speed, the process control system will fully use the medium load model G2; When the dispatch variable w ≥ 90% ID fan rated speed, the process control system will fully use the high load model G3. When the fan rated speed dispatch variable w is between 50% and 70%, the process control system uses 33.33% low load model, 66.67% medium load model and 0.0% high load model.

2. The ID fan optimization control method of aluminum hydroxide roasting furnace based on multi-model linear interpolation according to claim 1 is characterized in that: The process control system obtains the current control model by weighting through the following formula: G0=α1(w)G1+α2(w)G2+α3(w)G3 (1)

Images