Control system

A computer program product comprising processor executable instructions defining a program for use in a processor based apparatus for controlling a system having a first number of system inputs and a second number of system outputs, by outputting, at intervals, signals for updating the system inputs, the program defining:

• means for receiving one or more signals indicative of target values for the system outputs;
• mapping means for using steady state information that relates the system outputs to the system inputs under steady state conditions, to determine a number, equal to the first number, of steady state values for the system inputs on the basis of the target output values;
• means for receiving a number, equal to the second number, of signals indicative of the system outputs;
• initial value means for producing at least one set of initial values, wherein each set consists of a number, equal to the first number, of values for use as a basis for determining a set of update values, wherein the set of update values consists of a number, equal to the first number, of values for updating the system inputs;
• prediction means for using a model of the system to predict, for each set of initial values, a set of future values of the system outputs on the basis of past system input values, past system output values and a respective set of initial values;
• first differencing means for determining differences between the steady state values and each set of initial values to give first difference information;
• second differencing means for determining differences between the target values and each set of predicted future values of the system outputs to give second difference information;
• combining means for combining the first difference information with the second difference information to give combined difference information;
• determining means for determining the set of update values on the basis of the combined difference information and each set of initial values;
• and
• output means for outputting signals, indicative of the set of update values, for updating the system inputs.

A method of controlling a system having a first number of system inputs and a second number of system outputs, by outputting, at intervals, signals for updating the system inputs, the method comprising the steps of:

• receiving one or more signals indicative of target values for the system outputs;
• determining, on the basis of steady state information that relates the system outputs to the system inputs under steady state conditions, a first number of steady state values for the system inputs on the basis of the target output values;
• receiving a number, equal to the second number, of signals indicative of the system outputs;
• producing at least one set of initial values, wherein each set consists of a number, equal to the first number, of values for use as a basis for determining a set of update values, wherein the set of update values consists of a number, equal to the first number, of values for updating the system inputs;
• using a model of the system to predict, for for each set of initial values, a set of future values of the system outputs on the basis of past system input values, past system output values and a respective set of initial values;
• determining differences between the steady state values and each set of initial values to give first difference information;
• determining differences between the target values and each set of predicted future values of the system outputs to give second difference information;
• combining the first difference information with the second difference information to give combined difference information;
• determining the set of update values on the basis of the combined difference information and each set of initial values;
• and
• outputting signals, indicative of the set of update values, for updating the system inputs.