CLIAMS:WE CLAIM
1. An automated temperature control system for Decarburization and/or Annealing furnaces of silicon steel mill comprising
artificial intelligence module based on operating aspects including line speed of the steel in the furnace and the steel physical parameters including steel grade to resolve real time control problem;
said artificial intelligence module activated an interference engine generating control parameters including predicted temperature routine for most appropriate thermal regime
Expert system having a PC based computing platform implementing for the decarburization zone and/or the annealing zone in the furnace;
said artificial intelligence activated interference engine disposed in bidirectional data communication with PLC based control system of the furnace for receiving the line speed measured by the PLC sensors and downloading the predicted temperature routine to the PLC based control system for regulating heater means in the decarburization zone and/or the annealing zone for controlling the temperature in the decarburization zone and/or the annealing zone according to said predicted temperature routine.

2. The automated temperature control system as claimed in claim 1, wherein the artificial intelligence module includes a Human Machine Interface with an interactive GUI screens for enabling user to feed the steel physical parameters, process visualization, rendering a view of actual zone temperatures as well as set temperatures in the decarburization zone and the annealing zone predicted by the system.

3. The automated temperature control system as claimed in anyone of claims 1 or 2, includes Level-I PLC sensors for measuring the actual line speed of the steel while it is continuously feed horizontally into the furnace.

4. The automated temperature control system as claimed in anyone of claims 1 or 2, wherein the bidirectional data communication with the PLC based control system involves OPC ( Object linking and embedding for Process Control) based interface for integrating different automation devices in the PLC based control system of the furnace with the Expert system;
said artificial intelligence module configures PID control modules of the PLC with the predicted temperature routines using synchronous data access at an interval of every 2 second by involving client server link of the PLC OPC server;
said artificial intelligence module reads various process data periodically at 2 second from the PLC like actual line speed, actual furnace temperature, coil parameters etc, the actual process data and predicted zone temperatures are periodically saved in a computer database to monitor the efficacy of control system;

5. The automated temperature control system as claimed in anyone of claims 1 to 4, wherein the interfacing engine predicts the optimum temperature routine for controlling the temperature in the decarburization zone and/or the annealing zone in the furnace to achieve desired magnetic properties in the steel which makes it for use in electromagnetic application.

6. The automated temperature control system as claimed in anyone of claims 1 to 5, wherein the optimum furnace temperature routine for Decarburization and Annealing of the steel sheet in the silicon steel mill selected by the rules and the knowledge base is adapted to indicate the temperatures in the decarburization zone to be maintained through seven zones and temperatures in the annealing area to be maintained through five zones in silicon steel mill.

7. The automated temperature control system as claimed in anyone of claims 1 to 6, wherein the interfacing engine comprises
artificial intelligence module including artificial intelligence based operating model developed by involving knowledge base created with historical standard furnace operating procedures and furnace operational data, said furnace operational data pertaining to the knowledge base is stored in a relational database tables created in the PC based computing platform;
set of rules framed based on standard operational practices of the steel mill.

8. The automated temperature control system as claimed in anyone of claims 1 to 7, wherein the interfacing engine is adapted to triggers the knowledge base and selects the fittest rule and works synergistically with the steel physical parameters to select most appropriate values of temperatures in the in different zones of the decarburization zone and/or the annealing zone.

9. The automated temperature control system as claimed in anyone of claims 1 to 8, wherein the control parameters including rules and the knowledge base are coupled through queries articulated in Structured Query Language (SQL) format.

10. The automated temperature control system as claimed in anyone of claims 1 to 9, wherein the knowledge base includes historical standard furnace operating procedures and furnace operational data for different steel grades in the production range of silicon steel Mill like C-330, C-350, C-400, C-470, C-530, C-600, C-700, C-800, C-900 etc. and the line speed range of 18 – 29 meter / second.

Dated this the 3rd day of November, 2014

Anjan Sen
Anjan Sen & Associates
(Applicants Agent)
IN/PA-199
,TagSPECI:FIELD OF THE INVENTION:
The present invention relates to controlling temperature in a Decarburization Annealing process in Silicon Steel Mill. In particular, the present invention is directed to develop an automated artificial intelligence module (Expert System) guided temperature control system for Decarburization of Annealing Furnaces of Tandem Annealing Lines in a Silicon Steel Mill. The present temperature control system advantageously controls the furnace zone temperatures during decarburization and annealing based on the steel sheet parameters like its width, thickness and steel grade as well as line speed without involving any human or operator’s intervention or assistance.

BACKGROUND:
Silicon Steel Mill (SSM) produces Cold Rolled Non Oriented (CRNO) electrical grade steels, which are widely used for motors, generators, alternators, ballasts, small transformers and a variety of other electromagnetic applications. Electrical steel is produced from a steel melt which is further hot rolled into finished strips. The finished strip is further subjected to decarburization i.e. reduction in carbon content present in the steel strips. Finally the strips are annealed to improve its mechanical properties. Both the above operations are crucial for achieving the desired magnetic properties, which makes the steel strip suitable for use in electromagnetic application.
For decarburization, the strip of steel sheet is continuously fed horizontally into a furnace. The furnace is a long electric radiant type, through which steel strips passes at varying line speeds. Temperature along the furnace length is controlled by varying power supply to the heating elements. A Schematic diagram of existing Decarburization Annealing (DA) furnace for Silicon Steel Mill is shown in the accompanying figure 1.
Inside the furnace different temperatures zones are maintained, where the steel sheets are heated to the desired temperature as they pass through it. The temperature is normally maintained between 700 to 850 OC under moistened synthetic gas (containing hydrogen and nitrogen) atmosphere. Decarburization occurs mainly through chemical reaction of humidity (H2O) in the furnace atmosphere, which causes removal of carbon atoms i.e. oxidation. The hydrogen present in the atmosphere has an important role in preventing the undesired oxidation of Iron. After decarburization, steel strip enters annealing zone. Annealing is extremely important for final quality of steel in terms of mechanical properties like hardness, ductility, formability, tensile strength etc. Annealing is accomplished as steel sheets are passed through different temperature zones in furnace under inert environment of Nitrogen and thermal conductive environment of Hydrogen. As evident from the accompanying figure 1, that in typical Silicon Steel Mill, temperature in decarburization zone is maintained through seven zones indicated as ‘Z01’ to ‘Z07’. Similarly temperature in annealing area is maintained through five zones indicated as ‘Z08’ to ‘Z12’. Before final exit, the steel sheets are dried in a drying furnace and finally a coating of organic layer is applied on it.
Both the decarburization as well as annealing is an important factor upon which final quality acceptance of electrical grade steel depends. However, both the process heavily depends on correct selection of all the furnace zone temperatures and line speed. The temperature in different zones actually depends on steel sheet parameters like its width, thickness and steel grade. The temperature to be maintained at different zones also depends on line speed. The co- relation can be expressed in following equations:
?d_i = fnj (g, w, s, l) (1)
?a_k = fnm (g, w, s, l) (2)
where ?d_i = Decarburization zone temperature; i = 1, 2….7; j = 1, 2….7; ?a_k = Annealing zone temperature; k = 8, 9….12; m = 8, 9….12; g = coil thickness or gauge, w = coil width, s = steel grade and l = line speed. However in actual condition, no exact co-relations are known, which can define the exact nature of the functions ‘fnj’ and ‘fnm’ in above equations. Further the mathematical co-relation between temperatures ?d_i, ?a_k and output steel quality parameters like final carbon content, hardness, ductility, formability, tensile strength etc. are also not known. These are the cases where concepts of artificial intelligence and more particularly an automated expert system can be practically implemented.
The known art of Decarburization Annealing Furnace temperature control is through DCS or PLC based process control system. The process operator manually enters the set temperatures of every furnace zones as per his experience and standard operating practice. The actual zone temperature as measured by thermocouples mounted on the furnace acts as process variable. Based upon above two parameters, individual PID (Proportional – Integral – Derivative) control function block in PLC / DCS actually controls the furnaces temperature by regulating the electric heaters in every zone.
The manual control system does not co-relate to an established scientific method and heavily depends on one person’s decision, which always may not be optimum. The manual system is very much labors intensive as operator have to manually change the set-points of all the zones, whenever a new coil with different steel grade is introduced. In manual operation it is impossible to keep-on changing the zone set temperatures, whenever the line speed changes. Further the operator does not always have correct co-relation between set temperature vis-à-vis steel grade and variable line speed.
Thus there has been a need for developing an automated system which will control the furnace zone temperatures during decarburization and annealing based on the steel sheet parameters like its width, thickness and steel grade without involving any human or operator’s intervention or assistances.

OBJECT OF THE INVENTION:
It is thus the principal object of the present invention is to develop a system which would be adapted to automatically predict and control temperature in a Decarburization and/or Annealing furnaces of silicon steel mill.
Another object of the present invention is to develop an automated temperature control system which would be adapted to predict an optimum temperature routine for the Decarburization zone and/or the Annealing zone in the furnace of silicon steel mill to achieve desired magnetic properties in the steel which makes it for use in electromagnetic application.
Yet another object of the present invention is to develop an automated temperature control system which would be adapted to predict the optimum temperature routine through Level-II automation Expert System based on line speed of the steel in the furnace and the steel physical parameters including steel grade.
Another object of the present invention is to develop an automated temperature control system which would be adapted to perform synergistically with PLC control system of the furnace and thereby control the electric heaters in the furnace to implement the optimum temperature routine.

SUMMARY OF THE INVENTION:
Thus according to the basic aspect of the present invention there is provided an automated temperature control system for Decarburization and/or Annealing furnaces of silicon steel mill comprising
artificial intelligence module (Expert System) based on operating aspects including line speed of the steel in the furnace and the steel physical parameters including steel grade to resolve real time control problem;
said artificial intelligence module activated an interference engine generating control parameters including predicted temperature routine for most appropriate thermal regime
Expert system having a PC based computing platform implementing for the decarburization zone and/or the annealing zone in the furnace;
said artificial intelligence activated interference engine disposed in bidirectional data communication with PLC based control system of the furnace for receiving the line speed measured by the PLC sensors and downloading the predicted temperature routine to the PLC based control system for regulating heater means in the decarburization zone and/or the annealing zone for controlling the temperature in the decarburization zone and/or the annealing zone according to said predicted temperature routine.
According to another aspect in the present automated temperature control system, the artificial intelligence module includes a Human Machine Interface with an interactive GUI screens for enabling user to feed the steel physical parameters, process visualization, rendering a view of actual zone temperatures as well as set temperatures in the decarburization zone and the annealing zone predicted by the system.
According to another aspect, the present automated temperature control system includes Level-0 instrumentation sensors for measuring the actual line speed of the steel while it is continuously feed horizontally into the furnace.
According to a further aspect in the present automated temperature control system, the bidirectional data communication with the PLC based control system involves OPC ( Object linking and embedding for Process Control) based interface for integrating different automation devices in the PLC based control system of the furnace with the Expert system;
said artificial intelligence module configures PID control modules of the PLC with the predicted temperature routines using synchronous data access at an interval of every 2 second by involving client server link of the PLC OPC server;
said artificial intelligence module reads various process data periodically at 2 second from the PLC like actual line speed, actual furnace temperature, coil parameters etc, the actual process data and predicted zone temperatures are periodically saved in a computer database to monitor the efficacy of control system;
According to yet another aspect in the present automated temperature control system, the interfacing engine predicts the optimum temperature routine for controlling the temperature in the decarburization zone and/or the annealing zone in the furnace to achieve desired electro-magnetic properties in the steel which makes it for use in electromagnetic application.
According to a further aspect in the present automated temperature control system, the optimum furnace temperature routine for Decarburization and Annealing of the steel sheet in the silicon steel mill selected by the rules and the knowledge base is adapted to indicate the temperatures in the decarburization zone to be maintained through seven zones and temperatures in the annealing area to be maintained through five zones in silicon steel mill.
According to another aspect in the present automated temperature control system, the interfacing engine comprises
artificial intelligence module including artificial intelligence based operating model developed by involving knowledge base created with historical standard furnace operating procedures and furnace operational data, said furnace operational data pertaining to the knowledge base is stored in a relational database tables created in the PC based computing platform;
set of rules framed based on standard operational practices of the steel mill.
According to a further aspect in the present automated temperature control system, the interfacing engine is adapted to triggers the knowledge base and selects the fittest rule and works synergistically with the steel physical parameters to select most appropriate values of temperatures in the in different zones of the decarburization zone and/or the annealing zone.
According to a further aspect in the present automated temperature control system, the control parameters including rules and the knowledge base are coupled through queries articulated in Structured Query Language (SQL) format.
According to yet another aspect in the present automated temperature control system, the knowledge base includes historical standard furnace operating procedures and furnace operational data for different steel grades in the production range of silicon steel Mill like C-330, C-350, C-400, C-470, C-530, C-600, C-700, C-800, C-900 etc. and the line speed range of 18 – 29 meter / second.

BRIEF DESCRIPTION OF THE ACCOMPANYING FIGURES:
Fig 1: Schematic Diagram of DA furnace at Silicon Steel Mill
Fig 2: Architecture of Expert System based Control System
Fig 3: Main Human machine Interface Screen of Control system
Fig 4: Typical Process Trend depicting precision control

DETAILED DESCRIPTION OF THE INVENTION WITH REFERENCE TO THE ACCOMPANYING FIGURES
As stated herein above the present invention discloses an integrated expert system guided automatic temperature control system for implementing in a Decarburization and Annealing Furnaces of Silicon Steel Mill. The present System selects the optimum zone temperatures or furnace thermal regime as would have been done by a group of experts with an objective to attain desired steel quality.

The system of the present invention involves Artificial Intelligence based operating model to resolve a real time control problem. The present Expert System’s Artificial Intelligence based knowledge base in the inference engine takes most suitable decision on the controlling of temperature in a Decarburization and Annealing furnace, as would have been taken by experts. In view of this, an Expert System was developed for optimum selection of temperature routine for Decarburization Annealing furnaces of Tandem Annealing Line in a Silicon Steel Mill.

The knowledge base of the Artificial Intelligence module is created with historical standard furnace operating procedures and furnace operational data which for Different steel grades in the production range of Silicon Steel Mill like C-330, C-350, C-400, C-470, C-530, C-600, C-700, C-800, C-900 etc. and line speed range of 18 – 29 meter / second.

Relational database tables were created to store data pertaining to knowledge base. The actual line speed as communicated by Level-I PLC is being used for prediction. The other parameters like steel grade, sheet gauge and sheet width are manually entered by line operators in a specially designed Graphics User Interface (GUI) screen. Based on above data, Expert System inference engine triggers knowledge base and expert rules to select the most appropriate thermal regime. Several rules have been framed in the inference engine based on plant standard operational practices. Based on this knowledge base and selecting the fittest rule, expert system predicts optimum values of decarburization zone set temperatures ?d_1, ?d_2, ?d_3, ?d_4, ?d_5, ?d_6 and ?d_7 [equations (1) above].

The system also predicts optimum values of annealing zone set temperatures ?a_8, ?a_9, ?ad_10 and ?a_11 [equations (2) above].

The rules and knowledge base are coupled through queries articulated in Structured Query Language (SQL) format. For individual steel grades, different ranges of line speed were considered like speed less than 18 m/s, between 18-20, 20-22, 22-24, 24-26, 26-28, 28-29 and above 29. Further different coil thicknesses and widths were considered for different coil grades. The labyrinth conditions and nested queries framed for different combinations of steel grade, thickness, width and line speed were the core of Artificial Intelligence module, which works synergistically with rules and knowledge base to select most appropriate values of zone temperatures ?d_i and ?a_k.

Provision for fine tuning and amendment of knowledge base with suitable security protection was also incorporated in the system. The system also has features like incorporation of new steel grades, temperature ranges etc. in future. Architecture of Artificial Intelligence module based Control System is depicted in the accompanying figure 2.

Artificial Intelligence module downloads the selected zone temperatures to Level-I PLC and configures its PID control function blocks using OPC. The software modules of Level-II Expert System and Level-I PLC are under different programming platforms and different operating system (OS). Expert system was developed using Microsoft VC++ (MFC), while Rockwell Automation make Control-logix is used as PLC.

One of the challenges faced during program development was communication interfacing between Expert system’s Artificial Intelligence module and PLC. The problem was solved using OPC. OPC i.e. OLE (Object Linking and Embedding) for Process Control is an open interface for integrating different automation devices. It is basically based on Microsoft “COM / DCOM” and more recently “.Net” technology. OPC facilitates inter application linkage for data access, historical trends and alarm / event etc. It is based on Microsoft COM / DCOM base-interface ‘IUnknown’, which by virtue of its inheritance characteristic permits read / write data access between two applications written using different languages, developed under different platforms, under different operating systems and versions. Dedicated OPC client software was developed following the basic concepts of COM/DCOM. The data access with read – write interface was established with Rockwell Automation PLC OPC server named “RS-Linx OPC Server”. Using this client server link, artificial intelligence module configures PID control modules of PLC with selected set temperatures using synchronous data access at an interval of every 2 second. The artificial intelligence module also reads various process data periodically at 2 second from PLC like actual line speed, actual furnace temperature, coil parameters etc. The actual process data and predicted zone temperatures are periodically saved in a computer database to monitor the efficacy of control system.

The actual furnace temperature control is accomplished through PLC system. As PLC actually controls the furnace heating, its reliable operation is critical for maintaining un-interrupted plant operation. Keeping the above factor in consideration, it has been designed with high availability, fault tolerant and dual redundant configuration w.r.t. processor, communication and networking. The Level-0 automation tier comprises of field instrumentation which are connected with the PLC through various analog and digital input / output cards. The I/O’s are linked with processors through redundant deterministic network. Human Machine Interface (HMI) is connected with PLC using redundant industrial Ethernet. The instrumentation comprise of different zone temperature, synthesis gas flow & pressure, furnace pressure etc. PLC also takes care of various safety and other electrical interlocks required for furnace operation.

At regular interval of 2 second, artificial intelligence module configures the set-points of respective zone PID controller function block modules with the values of ?d_i and ?a_k [equations (1) and (2) above] and triggers it in “Auto” mode operation. The process variable of the controller is actual temperature signal received from thermocouple mounted on every zone. Based on the error signal and dead-band, PID controller generates output correction signal in the form of 0 – 100 %. This is converted to 4–20 mA by analog output cards, which is finally used as reference signal for thyristor controller of electric heater. Fig. 2 illustrates block diagram schematic of integrated process control.

A Human Machine Interface (HMI) was developed to facilitate the furnace operators with interactive GUI screens for process visualization, control and data entry. Fig. 3 depicts the main HMI screen, where operator feeds all the parameters related to steel sheet as mentioned in equation (1) and (2). It also renders a view of actual zone temperatures (PV) as well as set temperatures predicted by artificial intelligence module. The zone temperatures as selected by artificial intelligence module are shown in HMI as “TCR-SV” and actual process temperature are shown as “TCR-PV”. HMI also facilitates historical trends, alarm annunciations and other process graphics. A process trend depicting accurate control of zone temperature is shown in Fig. 4.

The expert system’s artificial intelligence module is implemented in a PC (personal computer) platform and software is written in VC++ (Microsoft Visual Studio). The PLC based control system is envisaged through a Rockwell Automation make Control Logix processor. The bidirectional data communication between expert system’s artificial intelligence module and Level–1 PLC has been achieved through an OPC (OLE for Process Control) client software developed in VC++ and is an integral part of expert system. Rockwell Automation make RS-Linx has been used as OPC server. As shown in Fig 3, operator enters coil grade. Based on this data and actual line speed as measured by sensor, the artificial intelligence module selects most suitable annealing cycle. The artificial intelligence module downloads the selected zone set temperatures to PLC, which finally regulates zone electrical heaters for controlling the temperature.

The novel features of the invention are explained below and protection shall be sought.
Development of rule based expert system’s artificial intelligence module comprising of knowledge base and inference rules for selection of optimum zone set temperature based on steel grade of coil and line speed.

Development of a PLC based software which facilitates actual control of furnace temperature as per regime decided by artificial intelligence module.
Interfacing of Expert System’s artificial intelligence module and PLC through a driver interface software based on OPC (OLE for Process Control)