CSense Advanced Process Control
Click
here to access full article.
APC Capabilities
CSense Systems has been specializing in APC solutions since 1998 and has assisted manufacturers and mining companies to implement APC solutions in order to get the most from their processes in the shortest possible time. CSense empowers its users to get results quickly by using the right technology. What follows will show how CSense can assist with the rapid development and deployment of APC solutions.
“APC is a proven technology that reduces process variability and inefficiency, improves product consistency, increases throughput by allowing operations to push constraints to the limits and achieve higher return on assets. Achieving and sustaining these benefits has not always been easy. It requires a sound strategy and adherence to best practices.” ARC Best Practices (Jan 2008).
APC Needs
When implementing an APC application, engineers must consider the needs of the system and these include:
- Ensuring that the real-time data is reliable
- Taking into account some of the available data pre-processing techniques
- Determining the available controller techniques and those that are available in CSense
Figure 1: The steps required for an APC implementation
Development and Deployment
In any APC implementation, simulation is critical because the processes under scrutiny normally deal with valuable material and assets and experimenting on live production plants is simply out of the question. The moment a control system goes live, the amount of tuning and experimentation needs to be minimized and that’s why it’s important that engineers have a flexible environment in which they can simulate the effect of their controllers and alter them without affecting live production.
CSense provides a visual development environment where there is no need for low-level programming. This is a modular, block diagram approach where users can drag and drop configurable components, which makes it more intuitive and quicker to develop a solution. The modular approach is important because the APC solution can be broken up into modules each of which addresses a specific functionality and also retains the intellectual property of its creator by being password protected.
- Reliability
- Open Architecture
- Maintainability
- Templates
- Top-down design
Data Validation
- Live data
- Rate of change
- Noise
- Fast Fourier Transforms (FFT)
Figure 2: Data validation functions
Pre-Processing
In the pre-processing step, CSense provides the following functionality:
- Resampling (up or down)
- Setting variable time delays
- Integration and derivation
- Scripting
- Rules
- Math
- Filtering
Controller
The control techniques that are available with CSense include:
1. State machine
2. General
3. PID blocks
4. Fuzzy control
5. Fuzzy PID
6. Model-predictive control
Modeling and Simulation
For modeling and simulation purposes, CSense provides a rich set of functions and figure 4 shows the various wave generators available as well as some simple blocks to create linear models for simple plant simulation purposes. Many processes can be approximated simply by first order delay with a time constant and a certain process gain. Or if that is not possible, perhaps the process can be approximated with a second order response.
Overview 
