According to the integration, the focus is on the design of the network and the application of expert experience. The simulation results show that the diagnostic method works well.

h Jie: Shi 1 Chang 4 on the main main signal and information processing research / / ww.Cnki.net At present, China's power industry has entered the development stage of large power grids and large units. The 300600MW generator set has become the main unit of the power grid. In order to prevent accidents in the operation of large-capacity generators, on the one hand, it is necessary to solve the design of the generator and the quality of the manufacturing process. On the other hand, it is necessary to improve the operation and maintenance level, improve the quality of the installation and maintenance of the unit, and carry out the purpose. Research on online monitoring and diagnostic technology to eliminate generator latent faults to prevent sudden accidents and reduce accident losses is necessary - at present, hydrogen cooled generator overheat alarm, radio frequency monitor, hydrogen leakage and hydrogen humidity have been developed in China. Monitoring, brushless excitation generator rotor current and temperature monitoring and other devices play a big role - but these devices are often operated separately, and some devices output data can only indicate indirect information of potential faults in various parts of the motor, and some It is difficult to determine the safety margin of operation - due to the large number of protection devices, it is difficult to deal with when the parameter cross-border alarm occurs. At the same time, because a large amount of information may be incomplete, inaccurate, or even contradictory, a slight accident or delay of the decision maker may cause a major economy. Loss and social impact - therefore automatic synthesis of this information is required The purpose of researching the online monitoring data fusion technology of generators is to apply information fusion technology to the field of industrial measurement and control to ensure safe, stable and economic operation of the unit.

1 Application of Data Fusion Technology in Generator Fault Diagnosis Synchronous generator is the basic equipment for producing electric energy and is the heart of the power grid. Its operational reliability directly affects the operation of the power grid and whether it can supply users safely and economically. For generators, not only in the design and manufacturing stages, but also to improve product quality and reliability as the primary task, and online monitoring and diagnosis after grid-connected operation has become an urgent issue. Its main purpose is to check the defects of the generator in the initial stage, in order to arrange the maintenance in a planned manner, thereby reducing the number of forced shutdowns and avoiding accidents; at the same time, prolonging the average time between failures of generators and shortening the average repair time and reducing maintenance costs And improve the availability of the generator. In order to ensure the safe and reliable operation of the generator, in the past decade or so, some countries in the world have carried out research on online monitoring and diagnostic technology, and gradually promoted the application. Generator failure monitoring and diagnostics have been listed as one of the central thesis of the SC-11 (Rotating Electric Machines) committee since the 80th annual meeting of the International Conference on Large Power Grids (CIGRE). In recent years, research and adoption of data fusion technology expert diagnostic systems have begun to make diagnostic techniques to a more advanced level.

Data fusion is an automated information synthesis processing technology developed and developed in the 1980s. It makes full use of the complementarity of multi-source data and the high-speed computing and intelligence of electronic computers to improve the quality of the resulting information. This technology was first widely used in the military and was quickly extended to areas such as automatic control and air traffic control medical diagnosis. It is a good application prospect to apply data fusion technology for fault diagnosis in generator online monitoring.

Generator operation monitoring data comes from different sensors. For example, on-line monitoring of stator winding insulation can obtain the RF level value of the RF monitor or the current indication value measured by the local overheat monitoring device.

The insulation deteriorates, the partial discharge increases, the RF level value is large, and when the insulation is partially overheated, the ionization current drops sharply, and the two data trends are opposite. Others such as the end vibration value of the hydrogen humidity data have specific meanings, reflecting the operating conditions of the various parts of the generator. It is difficult to extract the fault information of the essence, because the generator set has a large structure and the process is complicated and difficult to be. At present, the meter uses 8,10. It uses 8 inputs to reflect the expert experience knowledge. The input data is based on the comprehensive analysis of the output range of the specific online monitoring device, the normal operation data of the generator, and the empirical data of the unit under various conditions, and then normalized. It consists of 60 samples, of which 30 are for training and 30 are for testing. Among the three parameters of the Gaussian function network, each RBF center is determined by expert experience, and the weights of the e and output units are trained by the supervised learning method. The learning process adopts an iterative algorithm as follows: the minimum average distance of the samples; /=1, 2,... , N, N are the number of training samples, such as all Ek 2.4 The simulation operation result is 0.0235. After the iteration convergence, 30 test samples are input, and the diagnosis results of the stator and rotor conditions are obtained immediately. According to the output value of y1 to y6, if the threshold is 0.6 and the lower threshold is 0.3, then 29 samples are consistent with the expert experience knowledge, and one sample output y4 is only 0.556, and the error is large. In addition, when a new fault combination is input, the correct fusion result cannot be obtained. The new fault combination should be input as a training sample, retraining and learning, and improving the intelligence of the fusion network. Therefore, once a new sample is provided as an expert experience, it can be re-established. Training, the intelligence of the neural network can be continuously improved, simulating the human brain function, making the conclusion after data fusion more credible. Expert knowledge of domain experts is very important, directly affecting network intelligence and diagnostic decisions. Passwords must be set to ensure that the input is true expert experience.

The simulation results show that the reliability of the network diagnosis results is good, and the test samples with large individual errors can be re-entered into the network training as new samples after expert research. The new fault combination should be used as a new sample, re-enter the network, learn, and improve the intelligence of the converged network. After training, the network can correctly identify this type of fault combination. In the future decision-level integration, add more experience of human experts, such as considering generator capacity, process structure, commissioning time after overhaul, operating years, oil leakage and statistical rules of accident sites, and then combining RBF networks. The output results are subjected to fuzzy comprehensive evaluation to deal with inaccurate and uncertain conditions, and continuously improve the performance of the diagnostic decision system.

3 Conclusion Device diagnostic technology has a strong engineering background and has important practical value. Due to the combination of computer and monitoring technology, a new online monitoring system is constantly available, providing an effective technical means for predictive maintenance. The information fusion method more accurately identifies the fault conditions that cause the generator state to exceed the normal operating range, and realizes the comprehensive conclusion. The body-fitted grid of the twin-roll pool area is generated by the body-fitted coordinate transformation method and the speed proposed by Takuda is adopted. The model obtains the velocity field distribution of the molten pool area. Through the analysis of the relationship between enthalpy and latent heat, the temperature recovery source term processing method with steady-state temperature field with convection term is derived, and the heat transfer condition of the double-roller molten pool is carried out. The research results were studied and analyzed. The calculation results show that under the same cooling conditions, the casting temperature is increased to move the full solidification point of the casting belt toward the exit of the roll. For casting belts with a casting speed of 0.5 m/s, the pouring temperature of 14801500*C is suitable. Under the same cooling conditions, the casting speed of Q 5m/s is suitable. Too large or too small will affect the quality of the cast strip, and the cooling strength will match the casting speed. The calculation of the heat transfer calculation using the simple velocity field is the same as that obtained by the coupling of the flow field and the temperature field.

A rolling shutter door or roller shutter door is a type of door that consists of many horizontal slats hinge together and the door opens and closes by raising up or lowering down along a rolling shutter door track or guide rails.The rolling up door forming machine is automatic machine which can make different legnth as per your set.

The final product can be cut as per set length controller,and the length is automatically measured by encoder.When final product reached set length,machine will stop to cut the sheet automatically.Finish thhis action,the machine will run again continuously 

Metal Rolling Shutter Doors Forming Machine

Metal Rolling Shutter Doors Forming Machine,Door Slat Roll Forming Machine,Metal Door Frame Roll Forming Machine,0.8-1.2Mm Metal Thickness

Botou Golden Integrity Roll Forming Machine Co.,Ltd , https://www.jcxsteelrollformer.com