In this paper, the wireless communication network is established based on wireless communication equipment based on wireless network control technology. The wireless process control strategy for tank and high temperature furnace are designed, and the predictive function control (PFC⁃PID) and generalized open⁃loop response (GORC⁃PID) strategies are proposed respectively. Based on the requirement of task, the configuration of software and hardware for wireless industrial process system is achieved. On basis of this, the modelling of dynamic mathematic model of system and engineering configuration are completed. Finally, the industrial process control system based on wireless network is developed with the water tank and high temperature heating furnace of the National Local Joint Engineering Laboratory of Petrochemical Process Operation Optimization and Energy Saving Technology as the experimental device and the configuration software of Microwave Command Guidance System (MCGS) as the development and verification platform. The control results show that the developed wireless industrial process control system can control the level of tank and temperature of high temperature furnace effectively. It can avoid the drawback of wired communication at the industrial filed and has a very high industrial value.

In order to solve the problem of large interference of environmental noise during speech recognition, an isolated word recognition algorithm based on empirical mode decomposition (EMD) and dynamic time warping (DTW) is proposed. In this method, the EMD algorithm is used to decompose the speech signal with poor performance into several basic mode functions (IMF) and remove the interference and noise in the original signal. Then, based on the DTW algorithm, the short-time zero crossing rate and short-time energy are used to detect the endpoint detection of speech signal. After the speech feature parameters are extracted, the speech signal is matched with the reference template. Finally, the shortest path between the reference template and the template to be measured is used as the recognition result. The simulation results show that the proposed algorithm can improve the recognition efficiency of speech and the accuracy of recognition.

Aiming at the shortcomings of the conventional vision measurement method in the dynamic measurement of membrane structures, the target affects the local deformation of the film. The 3D dynamic measurement technology of membrane dot matrix projection is proposed. This technology uses a laser to project the target onto the membrane structure. Based on the OpenCV library（Open Source Computer Vision Library）, the 3D measurement algorithm is proposed to measure the 3D dynamic deformation of the membrane structure. In order to verify the 3D measurement accuracy, a standard ruler for accuracy inspection is designed. The 3D measurement accuracy of the membrane structure dot matrix projection measurement technology is tested in the principle verification experiment. The overall deformation and wave deformation of membrane structure are measured by 24 Hz frame rate photography and offline data processing. Accuracy verification experiments show that the technique can achieve the absolute accuracy of less than 0.30 mm and an accuracy of about 0.77 mm in the X, Y and Z directions, respectively, at the measurement distance of 1 m and the field of view of 1.3 m×0.8 m. At the same time, the dynamic measurement results can effectively replicate the dynamic deformation process of the membrane structure.

A new improved non-dominated sorting genetic algorithm (NSGA-II) is studied aiming at solving the low purity, high energy consumption problems existed in vinyl chloride rectification process. The method can be used to solve the multi-objective optimization problem of vinyl chloride rectification process. The multi-objective optimization function with the energy consumption and purity of vinyl chlorides based on considering the various constraints of the mechanism model and the actual production conditions were established through the sensitivity analysis for the main operating parameters such as the feeding position and reflux ratio of high and low boiling tower and so on. Finally, the objective function is solved by using the improved NSGA-II. Compared to the NSGA-II, the experimental results show that the improved algorithm can get more uniform distribution of Pareto optimal solution set, which provides a strong support for the selection of parameters in the process of vinyl chloride distillation.

Multidimension 01 knapsack problem is a typical NonDeterministic Polynomial problem. In view that ant colony algorithm solving often have local optimum and slow convergence speed problems, this article propose an improved ant colony algorithm. Based on the introduction of leap frog algorithm clustering thought, the ants have been developed into two groups with different culture, which improves the global search ability. The greedy strategy modified formula is applied in order to improve the accuracy of calculation. The formula is further simplified by roulette algorithm. Simulation results indicated that the modified algorithm improves the accuracy and global search ability.

The traditional envelope spectrum method does not accurately extracting characteristic problems in rolling bearing fault diagnosis, so a new energy and envelope spectrum based on a combination of timewavelet energy envelope spectrum analysis was developed. The various parts of rolling collected data the two methods were analyzed and compared, and show that the proposed method was better than the traditional envelope spectrum and could accurately extract the rolling bearing fault characteristic frequency.

 

 

Turbine rotor fault diagnosis is the key to ensuring the safe operation of the steam turbine. Vibration signal analysis is widely used in turbine rotor fault diagnosis. The wavelet packet analysis method was adopted to extract the vibration signal eigenvalue as the input of BP neural network, the nonlinear mapping relationship between signal features and fault type and realizing the fault diagnosis with BP neural network was established. The simulation results show that this method can effectively diagnosis turbine rotor fault.

In order to overcome the misclassification problems， multiway batch process monitoring based on the waveletbased multihidden Markov model tree (MHMT) was developed. MHMT can capture the clustering and persistence of the statistical characteristics for practical measured data. This approach provides less signal distortion and better understanding of the principal source of the system variability affecting the process. It solved the problem that discrete wavelet transform(DWT)can not obtain shiftinvariance，and then made simple modifications of the expansion structure， the method using stochastic model analysis the time domain expanded to timefrequency domain and extracted the main characteristics of the historical data， it is a good way to solve the multiway batch process monitoring problem. The proposed method was used to evaluate the industrial penicillin fermentation process data，the results clearly demonstrated the power and advantages of the proposed method in comparison with conventional MPCA method．