APC Technology

Fuzzy logic – a common, established AI library for easy-to-understand rule based reasoning logic, suitable for simple systems with minimal variables and dynamics. Fuzzy Logic uses membership functions and a fuzzy controller to define the degree of membership of an input value to a certain category. These functions map input values to membership degrees between 0 and 1, representing non-membership to full membership. These systems operate based on if-then rules expressed in a fuzzy manner. These rules connect input variables to output variables, allowing for nuanced reasoning. Instead of rigid true/false values, fuzzy logic acknowledges shades of gray. Statements can be partially true or false, reflecting real-world uncertainties, typical of most processing plants. ABB’s fuzzy logic control system library is intuitive and flexible, with multiple fuzzy library sizes for simple to complex block coding.

Model Predictive Control (MPC) using moving horizon estimation and use of optional Kalman filters, coupled with machine learning identification methods such as state space to develop dynamic models of the process. MPC relies on an internal dynamic model of the system to predict future behavior. It estimates the current state and calculates control actions over a finite horizon. At each time step, MPC minimizes a cost function by solving an optimization problem (where to drive the process that makes the most economic sense). The goal is to drive predicted outputs toward a reference trajectory while considering constraints (set by operations). MPC operates in a receding horizon manner. It computes control actions for a finite future window, applying only the first move to the system. MPC naturally handles complex systems which are typical of processing plants with interactions between inputs and outputs where it offers robustness against modeling errors and can handle time delays.

View, Scan, Track and Reporting PID tunings and performance.

ABB’s solution continuously visualizes and analyzes control loops, accelerating problem-solving. It identifies issues promptly, leading to improved control performance. By improving PID responses, it enhances availability, utilization, and efficiency. This translates to lower maintenance costs, optimized raw material usage, and reduced energy consumption.