Soft Measurement Research Of Methanol Synthesis Based On Improved Cloud Particle Swarm Optimization Algorithm With Neural Network

Cloud model theory is developing rapidly in recent years, cloud model can organic combine randomness and fuzziness, and it has unique advantages in the transformation between qualitative and quantitative. To the concept of uncertainty, cloud model has very good expression ability, and it have been widely applied in many fields. The calculate amount of particle swarm algorithm is small, the operation is simple and has strong robustness, but it also has its own shortcomings, for example "premature convergence", and easy to fall into local optimal solution. Considering the advantages of cloud model and cloud the characteristics of randomness and stable tendency of droplets can effectively maintain the population diversity. In this paper, cloud model is introduced, then combined with particle swarm optimization (pso) to determine the inertia weight of algorithm, speed up the convergence speed of the algorithm, and improves the searching capability of the algorithm. Then the improved algorithm is combined with neural network, build a soft sensor model.Modeling plays a very important role in industrial production, in methanol production it’s multivariable, nonlinear and time-varying characteristics of chemical process, accurate and reliable model established for its process is particularly difficult, the development of advanced control and soft measurement technology of mature offers the potential for this purpose. Based on the cloud particle swarm optimization (cpso) algorithm and neural network, soft measurement model for methanol production in the process of purification engineering modeling, synthesis, transformation is built, and use the actual production data to validate the precision of the model, results show that the model can meet the practical needs.To provide enough data for methanol production information, real-time monitoring the production situation of methanol, using Visual C++and kingview software to develop the online operating software for methanol production, the software use the soft measurement modeling method introduced in this paper, to establish the soft measurement module, can realize the observed quantity forecast estimates, so as to guide the rational operation, the operator to improve the production efficiency and increase enterprise benefit.