阅读说明：本技术 一种热镀锌机组锌层厚度控制方法 (Method for controlling thickness of zinc layer of hot galvanizing unit ) 是由 夏志 廖砚林 柳会梅 周云根 熊俊伟 于 2019-10-21 设计创作，主要内容包括：一种热镀锌机组锌层厚度控制方法,所述方法包括：采集热镀锌机组用于计算锌层厚度预测值的工艺参数,其中,所述工艺参数包括带钢速度、气刀高度、刀唇到带钢距离、气刀压力、刀唇间隙和带钢厚度；判断当前工况变化情况,基于工艺参数计算出当前工况变化情况下对应的所有可能的锌层厚度预测值,找出与锌层厚度设定值最接近的一组锌层厚度预测值,基于该组锌层厚度预测值对应的工艺参数对气刀对应参数进行调节控制。本发明通过计算出的刀唇到带钢距离以及气刀压力对气刀的对应参数进行调节控制,从而缩小锌层厚度与锌层厚度设定值之间的偏差,能够满足多种工况的锌层厚度控制要求,提高热镀锌机组锌层厚度控制精度与均匀性。(A method for controlling the thickness of a zinc layer of a hot galvanizing unit comprises the following steps: collecting technological parameters of a hot galvanizing unit for calculating a predicted value of the thickness of a zinc layer, wherein the technological parameters comprise strip steel speed, air knife height, distance from a knife lip to the strip steel, air knife pressure, a knife lip gap and strip steel thickness; and judging the change condition of the current working condition, calculating all possible zinc layer thickness predicted values corresponding to the change condition of the current working condition based on the process parameters, finding out a group of zinc layer thickness predicted values closest to the zinc layer thickness set value, and adjusting and controlling parameters corresponding to the air knife based on the process parameters corresponding to the group of zinc layer thickness predicted values. According to the invention, the corresponding parameters of the air knife are adjusted and controlled through the calculated distance between the knife lip and the strip steel and the air knife pressure, so that the deviation between the thickness of the zinc layer and the set value of the thickness of the zinc layer is reduced, the thickness control requirements of the zinc layer under various working conditions can be met, and the thickness control precision and uniformity of the zinc layer of the hot galvanizing unit are improved.)

1. A method for controlling the thickness of a zinc layer of a hot galvanizing unit is characterized by comprising the following steps:

Step 1, collecting technological parameters of a hot galvanizing unit for calculating a predicted value of the thickness of a zinc layer, wherein the technological parameters comprise strip steel speed, air knife height, distance from a knife lip to the strip steel, air knife pressure, a knife lip gap and strip steel thickness;

And 2, judging the change condition of the current working condition, calculating all possible zinc layer thickness predicted values corresponding to the change condition of the current working condition based on the process parameters, finding out a group of zinc layer thickness predicted values closest to the zinc layer thickness set value, and adjusting and controlling parameters corresponding to the air knife based on the process parameters corresponding to the group of zinc layer thickness predicted values.

2. the method for controlling the thickness of the zinc layer of the hot galvanizing unit according to claim 1, wherein the step 2 specifically comprises the following steps:

when the working condition changes greatly, the distance from the knife lip to the strip steel and the air knife pressure of each time point are obtained, the distance from a plurality of groups of knife lips to the strip steel and the pressure of a plurality of groups of air knives are obtained, the predicted value of the thickness of a zinc layer is calculated based on the distance from the plurality of groups of knife lips to the strip steel, the arbitrary pairwise combination of the pressure of the plurality of groups of air knives and the strip steel speed, the air knife height, the gap between the knife lips and the strip steel at the current moment, the predicted value of the thickness of the zinc layer is obtained, a group of predicted value of the thickness of the zinc layer closest to the set value of the thickness of the zinc layer is found out, and the corresponding parameter of;

when the working condition changes little, the air knife pressure of each time point is obtained, a plurality of groups of air knife pressures are obtained, a predicted value of the thickness of the zinc layer is calculated based on the air knife pressure of each group, the distance from the knife lip to the strip steel, the strip steel speed, the air knife height, the gap between the knife lip and the strip steel thickness at the current moment, a plurality of groups of predicted values of the thickness of the zinc layer are obtained, a group of predicted values of the thickness of the zinc layer closest to the set value of the thickness of the zinc layer is found, and the air knife corresponding parameters are adjusted and controlled based on the.

3. The method for controlling the thickness of the zinc layer of the hot galvanizing unit according to claim 2, wherein the working condition change condition is judged according to the strip steel speed change rate and the strip steel thickness change rate; when the strip steel speed change rate is greater than a strip steel speed change rate preset value s or the strip steel thickness change rate is greater than a strip steel thickness change rate preset value t, judging that the working condition change is large: and when the strip steel speed change rate is smaller than the strip steel speed change rate preset value s or the strip steel thickness change rate is smaller than the strip steel thickness change rate preset value t, judging that the working condition change is small.

4. The method for controlling the thickness of the zinc layer of the hot galvanizing unit according to claim 2, further comprising the following steps:

When the working condition is unchanged, a zinc layer thickness measuring instrument is used for measuring a zinc layer thickness measuring value, the zinc layer thickness measuring value is used as a zinc layer thickness predicted value at the current moment, the air knife pressure is reversely calculated based on the zinc layer thickness predicted value, the distance from a knife lip to strip steel, the strip steel speed, the air knife height, the knife lip gap and the strip steel thickness at the current moment, and the air knife corresponding parameter is adjusted and controlled based on the air knife pressure as a target value.

5. The method for controlling the thickness of the zinc layer of the hot galvanizing unit according to the claim 1, wherein the predicted value of the thickness of the zinc layer is calculated based on a zinc layer thickness prediction neural network model after learning training, the input parameters of an input layer of the zinc layer thickness prediction neural network model are the speed of strip steel, the height of an air knife, the distance from a knife lip to the strip steel, the pressure of the air knife, the gap between the knife lip and the thickness of the strip steel, and after the input parameters are converted through a hidden layer, the output parameters of an output layer are the predicted value of the thickness of the zinc layer;

when the working condition changes greatly, the input parameters of the zinc layer thickness prediction neural network model are any two-two combination of the strip steel speed, the air knife height, the knife lip gap, the strip steel thickness, the distance from the knife lip to the strip steel at all the moments and the air knife pressure at all the moments, and a group of zinc layer thickness prediction values are output aiming at the combination of the distance from each knife lip to the strip steel and the air knife pressure, so that a plurality of groups of zinc layer thickness prediction values are obtained;

when the working condition changes little, the input parameters of the zinc layer thickness prediction neural network model are the strip steel speed, the air knife height, the knife lip gap, the strip steel thickness, the distance from the knife lip to the strip steel and the air knife pressure at any moment, and a group of zinc layer thickness prediction values are output according to the air knife pressure at each moment, so that a plurality of groups of zinc layer thickness prediction values are obtained.

6. the method for controlling the thickness of the zinc layer of the hot galvanizing unit according to claim 5, further comprising the following steps: and performing learning training on the zinc layer thickness prediction neural network model, wherein sample data of the learning training is measured values of the strip steel speed, the air knife height, the distance from the knife lip to the strip steel, the air knife pressure, the knife lip gap, the strip steel thickness and the zinc layer thickness acquired at each moment, input sample data of the learning training is measured values of the strip steel speed, the air knife height, the distance from the knife lip to the strip steel, the air knife pressure and the knife lip gap acquired at each moment, and output sample data of the learning training is measured values of the zinc layer thickness at corresponding moments.

7. The method for controlling the thickness of the zinc layer of the hot galvanizing unit according to claim 6, further comprising rejecting abnormal data in the sample data by using a threshold value, an angle distance method and a robust regression method to ensure the reliability of the sample data.

Technical Field

The invention relates to the technical field of cold-rolled strip steel hot galvanizing, in particular to a method for controlling the thickness of a zinc layer of a hot galvanizing unit.

Background

on a hot galvanizing production line, an important technical index for measuring the quality of products is the thickness and the uniformity of a coating. The coating is too thick, which affects the spot weldability, adhesiveness and powdering resistance of the coating, and wastes raw materials such as zinc ingots; the coating is too thin, which affects the corrosion resistance of the product and is generally not acceptable to users. The zinc layer thickness control level directly affects the hot-dip galvanized sheet product quality, product cost and market competitiveness of the product.

The zinc layer thickness control model developed by Bao steel has a pressure priority control mode and a distance priority control mode, a zinc layer thickness control system developed by saddle steel takes air knife pressure as a main control quantity, pressure is regulated preferentially, when the pressure is saturated, the air knife distance is regulated to correct the thickness deviation of a coating, in practical application, the zinc layer thickness control system cannot meet the requirement of accurately controlling the thickness of a zinc layer with large working condition change, and an operator with abundant experience is required to operate an air knife to control the thickness of the zinc layer of a hot galvanizing unit on site.

Disclosure of Invention

in order to solve the problems in the prior art, the invention provides a method for controlling the thickness of a zinc layer of a hot galvanizing unit, which has the following specific scheme:

a method for controlling the thickness of a zinc layer of a hot galvanizing unit comprises the following steps:

Step 1, collecting technological parameters of a hot galvanizing unit for calculating a predicted value of the thickness of a zinc layer, wherein the technological parameters comprise strip steel speed, air knife height, distance from a knife lip to the strip steel, air knife pressure, a knife lip gap and strip steel thickness;

And 2, judging the change condition of the current working condition, calculating all possible zinc layer thickness predicted values corresponding to the change condition of the current working condition based on the process parameters, finding out a group of zinc layer thickness predicted values closest to the zinc layer thickness set value, and adjusting and controlling parameters corresponding to the air knife based on the process parameters corresponding to the group of zinc layer thickness predicted values.

Further, step 2 specifically includes:

When the working condition changes greatly, the distance from the knife lip to the strip steel and the air knife pressure of each time point are obtained, the distance from a plurality of groups of knife lips to the strip steel and the pressure of a plurality of groups of air knives are obtained, the predicted value of the thickness of a zinc layer is calculated based on the distance from the plurality of groups of knife lips to the strip steel, the arbitrary pairwise combination of the pressure of the plurality of groups of air knives and the strip steel speed, the air knife height, the gap between the knife lips and the strip steel at the current moment, the predicted value of the thickness of the zinc layer is obtained, a group of predicted value of the thickness of the zinc layer closest to the set value of the thickness of the zinc layer is found out, and the corresponding parameter of;

when the working condition changes little, the air knife pressure of each time point is obtained, a plurality of groups of air knife pressures are obtained, a predicted value of the thickness of the zinc layer is calculated based on the air knife pressure of each group, the distance from the knife lip to the strip steel, the strip steel speed, the air knife height, the gap between the knife lip and the strip steel thickness at the current moment, a plurality of groups of predicted values of the thickness of the zinc layer are obtained, a group of predicted values of the thickness of the zinc layer closest to the set value of the thickness of the zinc layer is found, and the air knife corresponding parameters are adjusted and controlled based on the.

further, judging the working condition change condition through the strip steel speed change rate and the strip steel thickness change rate; when the strip steel speed change rate is greater than a strip steel speed change rate preset value s or the strip steel thickness change rate is greater than a strip steel thickness change rate preset value t, judging that the working condition change is large: when the strip steel speed change rate is smaller than the strip steel speed change rate preset value s or the strip steel thickness change rate is smaller than the strip steel thickness change rate preset value t, judging that the working condition change is small

Further, the method further comprises:

when the working condition is unchanged, a zinc layer thickness measuring instrument is used for measuring a zinc layer thickness measuring value, the zinc layer thickness measuring value is used as a zinc layer thickness predicted value at the current moment, the air knife pressure is reversely calculated based on the zinc layer thickness predicted value, the distance from a knife lip to strip steel, the strip steel speed, the air knife height, the knife lip gap and the strip steel thickness at the current moment, and the air knife corresponding parameter is adjusted and controlled based on the air knife pressure as a target value.

Further, a zinc layer thickness prediction value is calculated based on a zinc layer thickness prediction neural network model after learning training, input parameters of an input layer of the zinc layer thickness prediction neural network model are strip steel speed, air knife height, distance from a knife lip to strip steel, air knife pressure, a knife lip gap and strip steel thickness, and after the input parameters are converted through a hidden layer, output parameters of an output layer are the zinc layer thickness prediction value;

When the working condition changes greatly, the input parameters of the zinc layer thickness prediction neural network model are any two-two combination of the strip steel speed, the air knife height, the knife lip gap, the strip steel thickness, the distance from the knife lip to the strip steel at all the moments and the air knife pressure at all the moments, and a group of zinc layer thickness prediction values are output aiming at the combination of the distance from each knife lip to the strip steel and the air knife pressure, so that a plurality of groups of zinc layer thickness prediction values are obtained;

When the working condition changes little, the input parameters of the zinc layer thickness prediction neural network model are the strip steel speed, the air knife height, the knife lip gap, the strip steel thickness, the distance from the knife lip to the strip steel and the air knife pressure at any moment, and a group of zinc layer thickness prediction values are output according to the air knife pressure at each moment, so that a plurality of groups of zinc layer thickness prediction values are obtained.

further, the method further comprises: and performing learning training on the zinc layer thickness prediction neural network model, wherein sample data of the learning training is measured values of the strip steel speed, the air knife height, the distance from the knife lip to the strip steel, the air knife pressure, the knife lip gap, the strip steel thickness and the zinc layer thickness acquired at each moment, input sample data of the learning training is measured values of the strip steel speed, the air knife height, the distance from the knife lip to the strip steel, the air knife pressure and the knife lip gap acquired at each moment, and output sample data of the learning training is measured values of the zinc layer thickness at corresponding moments.

Further, the method further comprises the step of rejecting abnormal data in the sample data by using a threshold value, an angle distance method and a robust regression method, and the reliability of the sample data is ensured.

the invention has the following beneficial effects:

According to the method, a predicted value of the thickness of the zinc layer is calculated through technological parameters such as the speed of strip steel, the height of an air knife, the distance from a knife lip to the strip steel, the pressure of the air knife, the gap between the knife lip and the strip steel, the thickness of the strip steel and the like, the set value of the thickness of the zinc layer is taken as a target, the distance from the knife lip to the strip steel and the pressure of the air knife at each time point can be calculated on the basis of different working condition changes, the distance from the knife lip to the strip steel and the pressure of the air knife which are closest to the set value of the thickness of the zinc layer under each working condition change condition are calculated, and the corresponding parameters of the air knife are adjusted and controlled through the calculated distance from the knife lip to the strip steel and the pressure of the air knife, so that the deviation between the thickness of the zinc layer and.

Drawings

Fig. 1 is a flowchart of a method for controlling a thickness of a zinc layer of a hot galvanizing unit according to an embodiment of the present invention;

fig. 2 is a schematic diagram of a zinc layer thickness prediction neural network model provided in an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1, an embodiment of the present invention provides a method for controlling a thickness of a zinc layer of a hot dip galvanizing unit, where the method includes:

Step 1, collecting technological parameters of a hot galvanizing unit for calculating a predicted value of the thickness of a zinc layer, wherein the technological parameters comprise strip steel speed, air knife height, distance from a knife lip to the strip steel, air knife pressure, a knife lip gap and strip steel thickness;

And 2, judging the change condition of the current working condition, calculating all possible zinc layer thickness predicted values corresponding to the change condition of the current working condition based on the process parameters, finding out a group of zinc layer thickness predicted values closest to the zinc layer thickness set value, and adjusting and controlling parameters corresponding to the air knife based on the process parameters corresponding to the group of zinc layer thickness predicted values.

Wherein, step 2 specifically includes:

when the working condition changes greatly, the distance from the knife lip to the strip steel and the air knife pressure of each time point are obtained, the distance from a plurality of groups of knife lips to the strip steel and the pressure of a plurality of groups of air knives are obtained, the predicted value of the thickness of the zinc layer is calculated based on the distance from the plurality of groups of knife lips to the strip steel, the arbitrary pairwise combination of the pressure of the plurality of groups of air knives and the strip steel speed, the air knife height, the gap between the knife lips and the strip steel at the current moment, the predicted value of the thickness of the zinc layer is obtained, the predicted value of the thickness of the zinc layer closest to the set value of the thickness of the zinc layer is found out, and the corresponding parameter of the air.

When the working condition changes little, the air knife pressure of each time point is obtained, a plurality of groups of air knife pressures are obtained, a predicted value of the thickness of the zinc layer is calculated based on the air knife pressure of each group, the distance from the knife lip to the strip steel, the strip steel speed, the air knife height, the gap between the knife lip and the strip steel thickness at the current moment, a plurality of groups of predicted values of the thickness of the zinc layer are obtained, a group of predicted values of the thickness of the zinc layer closest to the set value of the thickness of the zinc layer is found, and the air knife corresponding parameters are adjusted and controlled based on the air.

when the working condition is unchanged, a zinc layer thickness measuring instrument is used for measuring a zinc layer thickness measuring value, the zinc layer thickness measuring value is used as a zinc layer thickness predicted value at the current moment, the air knife pressure is reversely calculated based on the zinc layer thickness predicted value, the distance from a knife lip to strip steel, the strip steel speed, the air knife height, the knife lip gap and the strip steel thickness at the current moment, and the air knife corresponding parameter is adjusted and controlled based on the air knife pressure as a target value.

Wherein the predicted value of the thickness of the zinc layer and the measured value of the thickness of the zinc layer correspond to values which are processed by an air knife.

The working condition change condition is judged according to the strip steel speed change rate and the strip steel thickness change rate; when the strip steel speed change rate is greater than a strip steel speed change rate preset value s or the strip steel thickness change rate is greater than a strip steel thickness change rate preset value t, judging that the working condition change is large: and when the strip steel speed change rate is smaller than the strip steel speed change rate preset value s or the strip steel thickness change rate is smaller than the strip steel thickness change rate preset value t, judging that the working condition change is small.

According to the method, a predicted value of the thickness of the zinc layer is calculated through technological parameters such as the speed of strip steel, the height of an air knife, the distance from a knife lip to the strip steel, the pressure of the air knife, the gap between the knife lip and the strip steel, the thickness of the strip steel and the like, the set value of the thickness of the zinc layer is taken as a target, the distance from the knife lip to the strip steel and the pressure of the air knife at each time point can be calculated on the basis of different working condition changes, the distance from the knife lip to the strip steel and the pressure of the air knife which are closest to the set value of the thickness of the zinc layer under each working condition change condition are calculated, and the corresponding parameters of the air knife are adjusted and controlled through the calculated distance from the knife lip to the strip steel and the pressure of the air knife, so that the deviation between the thickness of the zinc layer and.

The process parameters are the factors which have large influence on the thickness of a zinc layer by analyzing the main components of all production process parameters of the hot galvanizing unit, and the factors such as the speed of strip steel, the height of an air knife, the distance from the knife lip to the strip steel, the pressure of the air knife, the gap between the knife lip, the thickness of the strip steel and the like are selected as the main component factors through the main component analysis, so that the complexity of the model is reduced.

Preferably, a zinc layer thickness prediction value is calculated based on a zinc layer thickness prediction neural network model after learning training, input parameters of an input layer of the zinc layer thickness prediction neural network model are strip steel speed, air knife height, distance from a knife lip to strip steel, air knife pressure, knife lip gap and strip steel thickness, and after the input parameters are converted through a hidden layer, output parameters of an output layer are the zinc layer thickness prediction value;

When the working condition changes greatly, the input parameters of the zinc layer thickness prediction neural network model are any two-two combination of the strip steel speed, the air knife height, the knife lip gap, the strip steel thickness, the distance from the knife lip to the strip steel at all the moments and the air knife pressure at all the moments, and a group of zinc layer thickness prediction values are output aiming at the combination of the distance from each knife lip to the strip steel and the air knife pressure, so that a plurality of groups of zinc layer thickness prediction values are obtained;

When the working condition changes little, the input parameters of the zinc layer thickness prediction neural network model are the strip steel speed, the air knife height, the knife lip gap, the strip steel thickness, the distance from the knife lip to the strip steel and the air knife pressure at any moment, and a group of zinc layer thickness prediction values are output according to the air knife pressure at each moment, so that a plurality of groups of zinc layer thickness prediction values are obtained.

in the embodiment, the actual production data is taken as learning and training sample data and is close to the actual production by designing a zinc layer thickness prediction neural network model; the zinc layer thickness prediction module established by the system can continuously learn through training, so that very high precision, nonlinear mapping capability and generalization capability are achieved.

Preferably, the method further comprises learning and training a zinc layer thickness prediction neural network model, wherein the model can be selected from methods such as an artificial neural network, a nuclear partial least square method, a linear partial least square method and the like, the zinc layer thickness prediction neural network model designed on the basis of the artificial neural network method is shown in fig. 2, the model consists of three layers, the first layer is an input layer and receives parameter input, the second layer is a hidden layer and converts input data, the third layer is an output layer and outputs target parameters, in addition, a plurality of connections exist among neurons and correspond to different and constantly-modifiable weights, sample data of the learning and training is strip steel speed, air knife height, distance from a knife lip to strip steel, air knife pressure, knife lip gap, strip steel thickness and zinc layer thickness measured values acquired at each moment, and input sample data of the learning and training is strip steel speed, air knife height, knife lip distance, air knife pressure, knife, The height of the air knife, the distance between the knife lip and the strip steel, the pressure of the air knife and the gap between the knife lips, and the output sample data of the learning training are the measured values of the thickness of the zinc layer at the corresponding moment. The sample data for a portion of the training is shown in table 1:

TABLE 1

after training is finished, training and verifying the zinc layer thickness prediction artificial neural network on the basis of the collected zinc layer thickness measurement value sample data. And selecting 80% of data as sample data, 20% of data as verification data, selecting a Sigmoid function as a conversion function, training and verifying the zinc layer thickness prediction neural network model by adopting a BP algorithm, and if the error control is within an allowable range, considering that the training effect is good. The results are shown in Table 2 and meet the requirements on site.

TABLE 2

Preferably, the method further comprises rejecting abnormal data in the sample data by using a threshold value, an angle distance method and a robust regression method, and ensuring the reliability of the sample data.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.