# US 7,526,463 B2 · Neural network using spatially dependent data for controlling a web-based process

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 7,526,463 B2
- Document ID: US7526463B2
- Status: Granted
- Kind: B2
- Application number: 11/129,062
- Filed: 2005-05-13
- Published: 2007-01-04
- Granted: 2009-04-28
- Earliest claimed: 2005-05-13
- Assignees: Rockwell Automation Technologies, Inc.
- Inventors: L. Paul Collette, III; W. Douglas Johnson
- Web page: https://patentgrep.com/patent/US7526463B2

## CPC

- G05B 13/027: {using neural networks only}
- G06N 3/045: Combinations of networks
- G06N 3/0499: Feedforward networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G05B 2219/32193: Ann, neural base quality management
- G05B 2219/33021: Connect plural macrocircuits, neural network modules in a larger network
- G05B 2219/33027: Artificial neural network controller
- G06N 3/105: {Shells for specifying net layout}
- Y02P 90/02: Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- Y02P: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS

## Abstract

System and method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, e.g., a web/sheet-based process for producing a web/sheet-based product. Input data comprising a plurality of input data sets are provided to a neural network (analog or computer-based), each data set comprising values for one or more input parameters, each comprising a respective process condition or product property. The input data preserve spatial relationships of the input data. The neural network generates output data in accordance with the input data, the output data comprising a plurality of output data sets, each comprising values for one or more output parameters, each comprising a predicted process condition or product property. The output data preserve spatial relationships of the output data, which correspond to the spatial relationships of the input data. The output data are useable by a controller or operator to control the process.

## Claims

### Claim 1

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data, wherein providing the input data comprises one or more of:
synthesizing one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
synthesizing one or more product properties at each of one or more spatial positions on the product to generate the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
providing the output data to the controller or operator; and
the controller or operator controlling the process in accordance with the output data.

### Claim 3 (depends on 1)

The method of claim 1, wherein the process comprises a web based manufacturing process for producing a web product.

### Claim 4 (depends on 3)

The method of claim 3, wherein the web product comprises one or more of:
paper;
plastic;
polymer;
wood or organic composites;
particle-board;
sheet metal;
sheet composites;
laminates;
sheet glass;
fiber-glass;
textiles;
candy; or
gum.

### Claim 5 (depends on 1)

The method of claim 1, wherein the process conditions comprise one or more of:
temperature;
position;
pressure;
humidity;
voltage;
current;
flow;
speed;
rate;
feed properties; or
raw material properties.

### Claim 6 (depends on 1)

The method of claim 1, wherein the product properties comprise one or more of:
weight;
moisture content;
color;
strength;
composition;
flatness;
stiffness;
hardness;
texture;
thickness;
gloss;
runnability; or
printability.

### Claim 7 (depends on 1)

The method of claim 1, wherein said synthesizing comprises one or more of:
interpolating measured data;
extrapolating measured data; or
one or more models synthesizing the data based on measured data.

### Claim 8 (depends on 7)

The method of claim 7, wherein said modeling data is performed with a neural network configured to receive one or more measured data, and to generate one or more synthetic data in response to the one or more measured data.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
providing the output data to an operator; and
the operator manually controlling an actuator array to change a controllable process state in accordance with the output data, wherein the actuator array comprises a plurality of actuators, wherein the plurality of actuators have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 10 (depends on 1)

The method of claim 1, further comprising:
providing the output data to a controller;
the controller computing controller output data using the first output data as controller input data in place of measurement input data; and
the controller controlling an actuator array to change a controllable process state in accordance with the controller output data, wherein the actuator array comprises a plurality of actuators, wherein the plurality of actuators have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 11 (depends on 10)

The method of claim 10, wherein said providing the output data to a controller comprises:
providing the output data to a data network; and
the controller retrieving the output data from the data network.

### Claim 12 (depends on 11)

The method of claim 11, wherein the data network comprises a historical database.

### Claim 13 (depends on 10)

The method of claim 10, wherein said providing the input data to a neural network comprises:
providing the input data to a data network; and
the neural network retrieving the input data from the data network.

### Claim 14 (depends on 13)

The method of claim 13, wherein the data network comprises a historical database.

### Claim 15 (depends on 10)

The method of claim 10, further comprising:
performing said providing the input data to the neural network, said neural network generating output data in accordance with the input data, said providing the output data to the controller, said controller computing controller output data using the output as controller input data, and said controller controlling an actuator array to change a controllable process state in accordance with the controller output data, in an iterative manner to control production of the product with desired properties.

### Claim 16 (depends on 15)

The method of claim 15, wherein the input data for each iteration comprise an order stamp indicative of the order of the input data.

### Claim 17 (depends on 15)

The method of claim 15, wherein the order stamp comprises a time stamp indicative of the chronology of the input data.

### Claim 18 (depends on 15)

The method of claim 15, further comprising:
periodically updating the neural network based on measured process conditions and product properties.

### Claim 19 (depends on 1)

The method of claim 1, wherein the neural network comprises a plurality of sub-networks.

### Claim 20 (depends on 19)

The method of claim 19,
wherein each of the plurality of sub-networks corresponds to a respective data set in the plurality of data sets; and
wherein each input parameter in each data set corresponds to a respective input node of a sub-network.

### Claim 21 (depends on 1)

The method of claim 1, further comprising:
training the neural network using historical and synthesized process condition data, wherein the historical and synthesized process condition data has spatial relationships corresponding to the spatial relationships of the input data.

### Claim 22 (depends on 1)

The method of claim 1, wherein said providing input data to the neural network further comprises providing one or more single or average data values corresponding to one or more general process conditions or product properties.

### Claim 23 (depends on 22)

The method of claim 22, wherein said neural network generating output data further comprises generating one or more single or average data values corresponding to one or more predicted general process conditions or product properties.

### Claim 24 (depends on 23)

The method of claim 23, wherein said providing the output data to the controller further comprises providing the one or more single or average data values.

### Claim 25 (depends on 1)

The method of claim 1, wherein inputs and/or outputs of the neural network correspond to the spatial relationships of the input data.

### Claim 26 (depends on 1)

The method of claim 1,
wherein each input data set further comprises position information for the set of input parameters, wherein the position information indicates the spatial relationships of the input data; and
wherein each output data set further comprises position information for the one or more output parameters.

### Claim 27 (depends on 1)

The method of claim 1,
wherein the input data are provided to inputs of the neural network in a manner that preserves the spatial relationships of the input data;
wherein the output data are provided by outputs of the neural network in a manner that preserves spatial relationships of the output data; and
wherein the spatial relationships of the output data correspond to the spatial relationships of the input data.

### Claim 28 (depends on 1)

The method of claim 1, wherein the neural network comprises one or more of:
an analog neural network; or
a computer-based neural network.

### Claim 29 (depends on 1)

The method of claim 1, further comprising:
preprocessing the input data prior to said providing the data to the neural network.

### Claim 30 (depends on 29)

The method of claim 29, wherein said preprocessing the input data comprises one or more of:
removing and/or replacing unusable or invalid data with valid data; or
putting the data into a more usable form or format.

### Claim 31

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
a processor; and
a memory coupled to the processor that stores program instructions executable by the processor to implement a neural network, wherein the neural network comprises:
a plurality of inputs, operable to receive input data, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data, and wherein the input data are provided by one or more of:
synthesizing one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
synthesizing one or more product properties at each of one or more spatial positions on the product to generate the input data;
wherein the neural network is operable to generate output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the neural network further comprises:
a plurality of outputs, operable to output the output data, wherein the output data are useable by a controller or operator to control the process.

### Claim 32

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
means for providing input data to a neural network, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data, and wherein the means for providing input data comprise one or more of:
means for synthesizing one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data, or
means for synthesizing one or more product properties at each of one or more spatial positions on the product to generate the input data; and
means for the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process.

### Claim 33

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprises a plurality of input data sets, each comprising values for a set of input parameters and position information for the set of input parameters, wherein the position information indicates spatial relationships of the input data, wherein each input parameter in the set comprises a respective process condition or product property, wherein providing the input data comprises one or more of:
synthesizing one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
synthesizing one or more product properties at each of one or more spatial positions on the product to generate the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters and position information for the one or more output parameters, wherein each output parameter comprises a predicted process condition or product property, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process.

### Claim 34

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprises a plurality of data sets, each comprising values for a set of input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data are provided to inputs of the neural network in a manner that preserves spatial relationships of the input data, wherein providing the input data comprises one or more of:
synthesizing one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
synthesizing one or more product properties at each of one or more spatial positions on the product to generate the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, and wherein the output data are provided by the outputs of the neural network in a manner that preserves spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process.

### Claim 35 (depends on 1)

The method of claim 1, wherein providing the input data to the neural network comprises:
operating the process, wherein said operating comprises:
measuring process conditions and/or product properties at a plurality of positions to generate measurement data comprising measured process conditions and/or product properties, wherein the measurement data preserve spatial relationships among the measurement data;
wherein providing input data to the neural network comprises providing the measurement data to the neural network as an input data array, the method comprising:
computing controller output data using the output data as controller input data in place of measurement input data;
controlling an actuator array based on the controller output data to change a controllable process state using the actuators in accordance with the controller output data; and
performing said operating, said providing, said producing, said computing, and said controlling in an iterative manner to produce the product with desired properties.

### Claim 36 (depends on 1)

The method of claim 1, wherein providing the input data comprises:
measuring one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
measuring one or more product properties at each of one or more spatial positions on the product to generate the input data.

### Claim 37 (depends on 31)

The method of claim 31, wherein the .input data are provided by:
measuring one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
measuring one or more product properties at each of one or more spatial positions on the product to generate the input data.

### Claim 38 (depends on 32)

The method of claim 32, wherein the means for providing input data comprise:
means for measuring one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
means for measuring one or more product properties at each of one or more spatial positions on the product to generate the input data.

### Claim 39 (depends on 33)

The method of claim 33, wherein providing the input data comprises:
measuring one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
measuring one or more product properties at each of one or more spatial positions on the product to generate the input data.

### Claim 40 (depends on 34)

The method of claim 34, wherein providing the input data comprises:
measuring one or more process conditions at each of one or more spatial positions in a production line of the process to generate the input data; or
measuring one or more product properties at each of one or more spatial positions on the product to generate the input data.

### Claim 41

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the neural network comprises a plurality of sub-networks, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process.

### Claim 42

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process, the method further comprising:
training the neural network using historical and synthesized process condition data, wherein the historical and synthesized process condition data have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 43

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process, the method further comprising:
preprocessing the input data prior to providing the data to the neural network, wherein preprocessing the input data comprises one or more of:
removing and/or replacing unusable or invalid data with valid data; or
putting the data into a more usable form or format.

### Claim 44

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
a processor; and
memory coupled to the processor that stores program instructions executable by the processor to implement a neural network, wherein the neural network comprises:
a plurality of inputs, operable to receive input data, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data, wherein the neural network comprises a plurality of sub-networks;
wherein the neural network is operable to generate output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the neural network further comprises:
a plurality of outputs, operable to output the output data, wherein the output data are useable by a controller or operator to control the process.

### Claim 45

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
a processor; and
a memory coupled to a processor that stores program instructions executable by the processor to implement a neural network, wherein the neural network comprises:
a plurality of inputs, operable to receive input data, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data;
wherein the neural network is operable to generate output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the neural network further comprises:
a plurality of outputs, operable to output the output data, wherein the output data are useable by a controller or operator to control the process;
wherein the neural network is operable to be trained using historical and synthesized process condition data, wherein the historical and synthesized process condition data have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 46

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
a processor; and
a memory coupled to a processor that stores program instructions executable by the processor to implement a neural network, wherein the neural network comprises:
a plurality of inputs, operable to receive input data, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data;
wherein the neural network is operable to generate output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the neural network further comprises:
a plurality of outputs, operable to output the output data, wherein the output data are useable by a controller or operator to control the process;
wherein the program instructions are further executable to preprocess the input data prior to providing the data to the neural network, wherein preprocessing the input data comprises one or more of:
removing and/or replacing unusable or invalid data with valid data; or
putting the data into a more usable form or format.

### Claim 47

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
means for providing input data to a neural network, wherein the neural network comprises a plurality of sub-networks, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data; and
means for the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data; and wherein the output data are useable by a controller or operator to control the process.

### Claim 48

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
means for providing input data to a neural network, wherein the input data comprise a plurality of input data sets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data;
means for the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data; and wherein the output data are useable by a controller or operator to control the process; and
means for training the neural network using historical and synthesized process condition data, wherein the historical and synthesized process condition data have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 49

A system for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the system comprising:
means for providing input data to a neural network, wherein the input data comprise a plurality of input datasets, each comprising values for a set of one or more input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data preserve spatial relationships of the input data;
means for the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, wherein the output data preserve spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data; and wherein the output data are useable by a controller or operator to control the process; and
means for preprocessing the input data prior to providing the data to the neural network, wherein preprocessing the input data comprises one or more of:
removing and/or replacing unusable or invalid data with valid data; or
putting the data into a more usable form or format.

### Claim 50

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the neural network comprises a plurality of sub-networks, wherein the input data comprises a plurality of input data sets, each comprising values for a set of input parameters and position information for the set of input parameters, wherein the position information indicates spatial relationships of the input data, wherein each input parameter in the set comprises a respective process condition or product property; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters and position information for the one or more output parameters, wherein each output parameter comprises a predicted process condition or product property, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process.

### Claim 51

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprises a plurality of input data sets, each comprising values for a set of input parameters and position information for the set of input parameters, wherein the position information indicates spatial relationships of the input data, wherein each input parameter in the set comprises a respective process condition or product property; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters and position information for the one or more output parameters, wherein each output parameter comprises a predicted process condition or product property, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process, the method further comprising:
training the neural network using historical and synthesized process condition data, wherein the historical and synthesized process condition data have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 52

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprises a plurality of input data sets, each comprising values for a set of input parameters and position information for the set of input parameters, wherein the position information indicates spatial relationships of the input data, wherein each input parameter in the set comprises a respective process condition or product property; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters and position information for the one or more output parameters, wherein each output parameter comprises a predicted process condition or product property, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process, the method further comprising:
preprocessing the input data prior to providing the data to the neural network, wherein preprocessing the input data comprises one or more of:
removing and/or replacing unusable or invalid data with valid data; or
putting the data into a more usable form or format.

### Claim 53

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the neural network comprises a plurality of sub-networks, wherein the input data comprises a plurality of data sets, each comprising values for a set of input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data are provided to inputs of the neural network in a manner that preserves spatial relationships of the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, and wherein the output data are provided by the outputs of the neural network in a manner that preserves spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process.

### Claim 54

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprises a plurality of data sets, each comprising values for a set of input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data are provided to inputs of the neural network in a manner that preserves spatial relationships of the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, and wherein the output data are provided by the outputs of the neural network in a manner that preserves spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable by a controller or operator to control the process, the method further comprising:
training the neural network using historical and synthesized process condition data, wherein the historical and synthesized process condition data have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 55

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
providing input data to a neural network, wherein the input data comprises a plurality of data sets, each comprising values for a set of input parameters, wherein each input parameter in the set comprises a respective process condition or product property, and wherein the input data are provided to inputs of the neural network in a manner that preserves spatial relationships of the input data; and
the neural network generating output data in accordance with the input data, wherein the output data comprise a plurality of output data sets, each comprising values for a set of one or more output parameters, each output parameter comprising a predicted process condition or product property, and wherein the output data are provided by the outputs of the neural network in a manner that preserves spatial relationships of the output data, and wherein the spatial relationships of the output data correspond to the spatial relationships of the input data;
wherein the output data are useable a controller or operator to control the process, the method further comprising:
preprocessing the input data prior to providing the data to the neural network, wherein preprocessing the input data comprises one or more of:
removing and/or replacing unusable or invalid data with valid data; or
putting the data into a more usable form or format.

### Claim 56

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
operating the process, wherein said operating comprises:
measuring process conditions and/or product properties at a plurality of positions to generate measurement data comprising measured process conditions and/or product properties, wherein the measurement data preserve spatial relationships among the measurement data;
providing the measurement data to a neural network as an input data array, wherein the neural network comprises a plurality of sub-networks;
the neural network producing output data in response to the measurement data, wherein the output data comprise predicted process conditions and/or product properties, and wherein the output data have spatial relationships that correspond to the spatial relationships of the measurement data;
computing controller output data using the output data as controller input data in place of measurement input data;
controlling an actuator array based on the controller output data to change a controllable process state using the actuators in accordance with the controller output data; and
performing said operating, said providing, said producing, said computing, and said controlling in an iterative manner to produce the product with desired properties.

### Claim 57

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
operating the process, wherein said operating comprises:
measuring process conditions and/or product properties at a plurality of positions to generate measurement data comprising measured process conditions and/or product properties, wherein the measurement data preserve spatial relationships among the measurement data;
providing the measurement data to a neural network as an input data array;
the neural network producing output data in response to the measurement data, wherein the output data comprise predicted process conditions and/or product properties, and wherein the output data have spatial relationships that correspond to the spatial relationships of the measurement data;
computing controller output data using the output data as controller input data in place of measurement input data;
controlling an actuator array based on the controller output data to change a controllable process state using the actuators in accordance with the controller output data; and
performing said operating, said providing, said producing, said computing, and said controlling in an iterative manner to produce the product with desired properties, the method further comprising:
training the neural network using historical and synthesized process condition data, wherein the historical and synthesized process condition data have spatial relationships corresponding to the spatial relationships of the input data.

### Claim 58

A method for controlling a process with spatially dependent conditions for producing a product with spatially dependent properties, the method comprising:
operating the process, wherein said operating comprises:
measuring process conditions and/or product properties at a plurality of positions to generate measurement data comprising measured process conditions and/or product properties, wherein the measurement data preserve spatial relationships among the measurement data;
providing the measurement data to a neural network as an input data array;
the neural network producing output data in response to the measurement data, wherein the output data comprise predicted process conditions and/or product properties, and wherein the output data have spatial relationships that correspond to the spatial relationships of the measurement data;
computing controller output data using the output data as controller input data in place of measurement input data;
controlling an actuator array based on the controller output data to change a controllable process state using the actuators in accordance with the controller output data; and
performing said operating, said providing, said producing, said computing, and said controlling in an iterative manner to produce the product with desired properties, the method further comprising:
preprocessing the input data prior to providing the data to the neural network, wherein preprocessing the input data comprises one or more of:
removing and/or replacing unusable or invalid data with valid data; or
putting the data into a more usable form or format.