Making Industrial Processes More Cost Efficient
Coal fired power plants should be operated in such a way that the emissions are kept clearly below desired limits and the combustion efficiency is as high as can be achieved. This requires a lot of quantitative knowledge of the effects of the process variables and different fuel characteristics on the emissions and efficiency.Read More
An engineer working with secondary coating of optical fibre cables wants to minimise shrinkage, maintain excess fibre length below a certain limit, have the tensile modulus within certain upper and lower limits, and prefers to have the production unit operating at 350 to 450 m/min.Read More
All industrial processes can be made more efficient.
Industrial processes are rarely operated anywhere close to the economic optimum. The reason for it is simple: it is just too complicated to find the optimum by traditional means. A large number of variables affect the consequences of a process or a series of processes. The consequences include quality variables, production rate, yield, frequency of rejects, raw material consumption, energy efficiency, etc. These consequences depend on a number of process variables as well as the amounts of materials fed into the process, or the composition of the material. The relations between composition and process variables, with end results of the process, are not very simple or linear. In addition, there are always upper and/or lower limits on many of the variables.
Our company develops advanced non-linear models from process data, often in combination with the knowledge of the processes. Once the models are developed, they are implemented in suitable software which sits in dedicated hardware. When used efficiently to operate the process, these systems can help to improve production economics, reduce energy, reduce raw material consumption, improve quality, reduce rejects, increase the production rate or yield. Any one parameter can be emphasized to give maximum output or a group of parameters can be selected to arrive at optimum levels.
processMax+ brings the domain of non-linear modelling to process indsutries including thermal power plants. This concept has given tangible benefits to process industries and has taken these one notch up in efficiency, productivity, energy or quality or a combination of these. These not only translate into higher profits, it helps to maintain a competitive edge over others ensuring long term sustainability and compliances of statutory guidelines. processMax+ is a definitive next step of industrial evolution.
Proponents of linear techniques argue that one can include quadratic terms in linear regression to account for nonlinearities. This is usually not done, and even if it is done, it is not efficient. Just as nature is not linear, it is not very quadratic either. Nature does not follow the simplicities which we try to fit it in. New techniques of nonlinear modelling allow us to approximate nonlinearities without specifying in detail the nonlinearities to be accounted for. They allow for free form nonlinearities, unlike linear and nonlinear regression methods.
Most steel plants are interested in ways to influence the sinter quality. Sinter quality consists of several variables, the measurements of many of which are time consuming. Mathematical models can provide an alternative to frequent measurements of such variables, as well as provide a means of influencing those quality variables. In some plants, sulphur dioxide emissions are also an issue, and they too can be taken into account.