A hydraulically controlled backing roller includes a mandrel, two cover plates which are installed at two ends of the mandrel through screws respectively, and two screwdown gears which are installed at two end portions of the mandrel respectively. The two screwdown gears are engaged with an output rack of a screwdown hydraulic cylinder, multiple saddle ring sets are sleeved on the mandrel at equal intervals, a saddle ring of each of the saddle ring sets is fixed with a frame through a fan-shaped plate, a backing bearing is provided between two adjacent saddle ring sets; an inner eccentric ring of the each of the saddle ring sets is driven to rotate by hydraulic driving, so that the mandrel has a deflection deformation, and the deflection deformation is transmitted to other adjacent rollers through the backing bearing.

A thickness meter detects a thickness of an electrode plate of a secondary battery at three or more points in a width direction of the electrode plate. A calculator calculates three feature amounts of a first deviation between the thickness target value and a thickness measurement value of a point closest to a first compression mechanism among the three or more points, a second deviation between the thickness target value and a thickness measurement value of a point closest to a second compression mechanism among the three or more points, and a secondary component of a thickness profile of the electrode plate from thickness measurement values at the three or more points and the thickness target value, and adaptively changes the setting values of the first compression mechanism, the second compression mechanism, a first bend mechanism, and a second bend mechanism on the basis of the three feature amounts.

A thickness meter detects a thickness of an electrode plate of a secondary battery at three or more points in a width direction of the electrode plate. A calculator calculates three feature amounts of a first deviation between the thickness target value and a thickness measurement value of a point closest to a first compression mechanism among the three or more points, a second deviation between the thickness target value and a thickness measurement value of a point closest to a second compression mechanism among the three or more points, and a secondary component of a thickness profile of the electrode plate from thickness measurement values at the three or more points and the thickness target value, and adaptively changes the setting values of the first compression mechanism, the second compression mechanism, a first bend mechanism, and a second bend mechanism on the basis of the three feature amounts.

Provided is an LSTM-based hot-rolling roll-bending force predicting method including the steps of acquiring final rolling data of a stand of a stainless steel rolling mill when performing a hot rolling process, and dividing the data into a training set traindata and a test set testdata; normalizing the traindata; building a matrix P; using a last row of the matrix P as a label of the training set, namely a true value; calculating and updating an output value and the true value of a network; after network training is completed, taking the last m output data of the LSTM network as an input at a next moment, and then obtaining an output of the network at the next moment, wherein the output is a predicted value of the roll-bending force at the next moment; repeating the steps until a sufficient number of prediction data is obtained; and comparing the processed data with the true value in the testdata to check the validity of the network.

Provided is an LSTM-based hot-rolling roll-bending force predicting method including the steps of acquiring final rolling data of a stand of a stainless steel rolling mill when performing a hot rolling process, and dividing the data into a training set traindata and a test set testdata; normalizing the traindata; building a matrix P; using a last row of the matrix P as a label of the training set, namely a true value; calculating and updating an output value and the true value of a network; after network training is completed, taking the last m output data of the LSTM network as an input at a next moment, and then obtaining an output of the network at the next moment, wherein the output is a predicted value of the roll-bending force at the next moment; repeating the steps until a sufficient number of prediction data is obtained; and comparing the processed data with the true value in the testdata to check the validity of the network.

In a roll press device, rolling is performed by sandwiching an electrode plate of a secondary battery to be continuously conveyed by a first pressure roller and a second pressure roller. A controller controls loads of a first compression mechanism, a second compression mechanism, a first bend mechanism, and a second bend mechanism on the basis of setting values calculated by a calculator. The calculator previously changes the setting values of the first compression mechanism and the second compression mechanism such that a thickness change of the electrode plate according to a speed change of a conveyance line of the electrode plate decreases.

In a roll press device, rolling is performed by sandwiching an electrode plate of a secondary battery to be continuously conveyed by a first pressure roller and a second pressure roller. A controller controls loads of a first compression mechanism, a second compression mechanism, a first bend mechanism, and a second bend mechanism on the basis of setting values calculated by a calculator. The calculator previously changes the setting values of the first compression mechanism and the second compression mechanism such that a thickness change of the electrode plate according to a speed change of a conveyance line of the electrode plate decreases.

A method of controlling flatness of a strip of rolled material in a production line including a hot rolling mill and at least one cold rolling mill, downstream of the hot rolling mill, the method including determining flatness data of the strip in one or more of the at least one cold rolling mill and/or following passing of the strip through one or more of the at least one cold rolling mill; determining a thickness profile target of the strip for the hot rolling mill based on the flatness data; and passing the strip through the hot rolling mill and adjusting the thickness of the strip based on the thickness profile target. A control system and a production line are also provided.

A method of controlling flatness of a strip of rolled material in a production line including a hot rolling mill and at least one cold rolling mill, downstream of the hot rolling mill, the method including determining flatness data of the strip in one or more of the at least one cold rolling mill and/or following passing of the strip through one or more of the at least one cold rolling mill; determining a thickness profile target of the strip for the hot rolling mill based on the flatness data; and passing the strip through the hot rolling mill and adjusting the thickness of the strip based on the thickness profile target. A control system and a production line are also provided.

A control device for a rolling mill apparatus including at least one rolling mill stand for rolling a metal plate includes: a detection signal acquisition part for receiving, from an edge crack sensor, a detection signal of an edge crack at an end portion of the metal plate in a plate width direction; and a rolling condition decision part for deciding a rolling condition for the rolling mill apparatus. The rolling condition decision part is configured to change, if the detection signal acquisition part receives the detection signal of the edge crack, the rolling condition for the rolling mill apparatus from a first rolling condition immediately before detection of the edge crack to a second rolling condition which is more capable of suppressing growth of the edge crack than the first rolling condition.