39results about How to "Improve quality control efficiency" patented technology

The invention discloses a direct quantitative calibration method for measuring hydrolyzable fluoride in sulfur hexafluoride gas, and belongs to the technical field of equipment detection. The direct quantitative calibration method for measuring hydrolyzable fluoride in sulfur hexafluoride gas comprises the following steps: taking easily hydrolyzable SOF2 standard gas as a quantitative reference, feeding the SOF2 standard gas into an absorbent solution, and completely dissolving the SOF2 standard gas in the absorbent solution, wherein the fluoride ion concentration of the absorbent solution is in direct proportion to the feeding volume of the SOF2 standard gas. According to the method, quantification of hydrolyzable fluoride in the sulfur hexafluoride gas directly traces to the source of national standard material, the difficulties that workload of hydrolyzable fluoride measuring experiment is large, the operation is complex, and the parallel test error is large are solved, the accuracy and stability of the experiment are ensured, and meanwhile, the efficiency of sulfur hexafluoride gas quality control is improved.

The invention provides a white vehicle body information management method and a system. The method comprises the following steps: according to the manufacturing process of a white vehicle body, an upper computer receives a production line welding parameter, a bolt torsion value, surface quality information and key size information of the white vehicle body in real time; a visual management interface is generated according to the production line welding parameter; a bolt is retightened according to the bolt torsion value; the surface quality information is associated with a corresponding net region of a white vehicle body view after net region division; and the key size information is associated with a corresponding size line of the white vehicle body view marked with the size line for display. The method can realize online information management, including the surface quality information, the key size, the production line welding parameter and the bolt torsion value of the white vehicle body, and prewarning, can improve stable, effective, comprehensive and accurate control of the quality of the white vehicle body, improves the quality control efficiency, reduces the staff waste, reduces the cost, and achieves the quality tracing and inquiring functions.

The invention relates to a method for simultaneously detecting a solvent and moisture in a polyurethane elastic fiber production solvent recovery system. The method comprises the following steps: (1) preparing a series of DMAC and H2O standard solutions of different concentrations, and selecting a sufficient amount of representative calibration set samples; (2) determining a near-infrared spectral spectrogram of the calibration set samples, carrying out spectral pretreatment on a wave band, of 12,000-4000cm<-1>, of the spectrogram, then, respectively correlating with the content of DMAC and H2O in a standard sample, and establishing a DMAC content calibration model and a H2O content calibration model by adopting a partial least square method; (3) determining a near-infrared spectral spectrogram of a sample to be detected, treating the wave band, of 12,000-4000cm<-1>, of the obtained spectrogram by adopting the same pretreatment method as in the step (2), and substituting a light absorbance value, obtained after pretreatment, in the corresponding calibration model obtained in the step (2), thereby obtaining the content of DMAC in the sample to be detected. The method has the advantages that the analysis process is simple and rapid, the interference resistance is strong, the analysis is non-destructive, the sampling quantity is small, and the like.

The invention relates to the technical field of quality control of radiotherapy plans, in particular to a neural network prediction device for VMAT radiotherapy plans, which is used to solve the problem of single control object, poor performance and large errors in the quality control methods of radiotherapy plans in the prior art question. The invention includes a data preparation module and a model design module, which are used to establish a multi-branch neural network model. The input of the multi-branch neural network model is processed to obtain multi-dimensional features, and then the predicted gamma pass rate is output; the model training module is used to perform first Feature extraction, and regression calculation of the gamma pass rate, and then backpropagation, update the parameters of the model, and iterate multiple times until good model parameters are obtained; the model test module is used to output the predicted gamma pass rate after forward calculation; In the present invention, radiotherapy plans for more body parts can be accepted through the above steps, with smaller performance errors and better effects.

The present invention proposes a quality control method based on a multivariate Z-score quality control chart, comprising: step a: obtaining multiple detection items detected by the detection equipment, that is, the Z-scores of multiple detection variables; step b: establishing concentric circles Multivariate Z score quality control chart; step c: mark the Z scores of the multiple detection variables on the multivariate Z score quality control chart respectively; and step d: mark by the Z score of the multiple detection variables The location obtains the quality control status of each of the detected variables. Through the present invention, the controlled or out-of-controlled state of multiple detected variables can be visually observed by using a multivariate Z score quality control chart. In addition, the control of system stability can be realized through the Zs variable Z score quality control chart.