39results about How to "Few parameters involved" patented technology

The invention discloses a yawing greasing method and device and a computer readable storage medium. The yawing greasing method comprises the steps that if it is detected that a yawing bearing of a wind generating set has the greasing demand, greasing operation is conducted on the yawing bearing in the first-time yawing process of the wind generating set; and if a difference value between a maximumangle value and a minimum angle value of the greased zone of the yawing bearing is smaller than a preset greasing angle after first-time yawing is finished, greasing operation is conducted on the yawing bearing according to the actual cabin position in the next-time yawing process, the greased zone obtained after first-time yawing is finished and the preset greasing angle. By adopting the technical scheme in the embodiment, the yawing bearing can be lubricated in the set yawing process, and frequent halting of the wind generating set is avoided.

The invention discloses a color image light source color estimation method based on classification and correction. First, the edge features of the image are extracted from a group of images with known light source colors, and then the least squares method is used to learn to obtain the relationship between the edge features and the light source. Correction matrix, and then extract edge features from the test image to be processed and multiply with the correction matrix to obtain a rough light source estimate; then find a class of training images with similar characteristics to the test image to be processed by looking for K adjacent images in the feature space. , so as to relearn and obtain accurate light source estimation. The present invention involves few parameters, and because the extracted features are simple and few in number, it also has the characteristics of simple calculation and fast speed; in addition, the present invention is based on the learning method, so the processing effect is good, the accuracy is high, and it is very suitable for Occasions where the estimation accuracy of the light source color is relatively high.

The invention discloses a distributed optical fiber vibration signal feature extraction and identification method, which belongs to the field of optical fiber sensing signal processing. First, the space-time matrix signal of a vibration source is collected, the spatial column signal is extracted, and short-term signal units are divided to construct optical cable vibration event data. Then build, train and tune an improved mCNN model, and perform feature evaluation on the features extracted by the model during tuning until the model iteration reaches the optimum; then use the optimal mCNN model to extract the temporal structure feature vector at multiple scales in parallel , reorganized into short-term feature sequences in chronological order, and constructed a temporal structure feature sequence set; finally, an HMM model was constructed and trained, and an offline vibration event HMM model library was constructed as a classifier for vibration source identification. The invention solves the problems that the prior art cannot simultaneously extract the local structural features and time sequence features of the distributed optical fiber vibration signal, and the vibration source identification accuracy and model generalization ability are low.

The invention discloses a method for comparing the suitability of TBM excavation, comprising the following steps: S1: counting the total length l III ; S2: Statistics of the total length l IV ; S3: Statistics of the total length l V ; S4: According to l III , l IV , l V The suitability parameter Pg of the TBM equipment in the tunnel statistical section is calculated by correspondingly reducing the corresponding surrounding rock conditions and proportions. The conclusion of the invention is simple, and the results are easy to compare: the overall excavation suitability comparison between different tunnels of the same TBM model and different TBM models of the same tunnel can be carried out. The relatively small value indicates that the excavation adaptability of the TBM equipment under the tunnel is strong, otherwise it indicates that the equipment suitability is weak.

The invention discloses a neutral-point voltage control method of an NPC (neutral-point converter) type three-level inverter based on interval selection, belonging to the technical field of power electronics The method comprises the steps: selecting a phase which is maximum in an absolute value in the three phases as an adjusting phase, wherein a peak point of the adjusting phase is taken as a symmetrical point in the adjusting interval and the adjusting range is not greater than pi / 2; measuring the neutral-point voltage Unp in real time, comparing the Unp with a given value; determining the voltage offset amount according to the comparison result so as to adjust the modulating wave in the adjusting interval; and comparing the adjusted modulating wave with a carrier wave to output a PWM (pulse-width modulation) signal so as to control the work state of a switch tube of an inverter. With the adoption of the neutral-point voltage control method, fewer parameters are related, and only the neutral-point voltage needs to be detected, so that the method is simple in control structure and convenient to realize.

The invention discloses a rapid identification method for corrosion control mode of steel bar macro cells in marine concrete, and measures the corrosion potential E of cathode steel bars before macro cell corrosion occurs in concrete structures mi‑c Corrosion potential E of anodic steel bars before corrosion occurs in the macro cell mi‑a , calculate the macrocell potential difference ΔE between the cathode and anode steel bars corr 1. The demarcation relation of macro battery corrosion control mode is E mi‑c =0.5×(ΔE corr 1)-b, b is a constant, when b is taken as 382, ​​it is the boundary between the control mode of cathode steel bar and the control mode of mixed control mode, and when b is taken as 496, it is the boundary line between the control mode of mixed control mode and the control mode of anode steel bar. The rapid identification method of the present invention involves few electrochemical parameters, convenient detection and operation, and simple analysis process.

The invention discloses a method for evaluating the consistency of processing strength of shredded leaves in a drum drying process based on the heating process of shredded shreds. B. Obtain the inlet temperature of the shredded blades and the outlet temperature of the shredded blades according to the temperature distribution curve; measure the moisture content of the shredded blades at the outlet of the corresponding drum; C, calculate the processing intensity of the shredded blades in the drying process of the batch according to the above parameters and formula; D. Judging the consistency of shredded shreds processing strength based on the absolute value of the difference between the processing strength of shredded blades in the drying process of the target drum or batch and the processing strength of shredded blades in the drying process of the drum or batch to be tested. The invention realizes the transformation of the processing intensity index of drum-dried shredded leaves from qualitative analysis to quantitative analysis, effectively evaluates the processing effect of different process parameter groups of the same drum, the drum processing effect of different production lines, etc., and has strong guiding significance for the actual production process .

The invention provides a classification method, device and storage device of hyperspectral remote sensing images, a classification method of hyperspectral remote sensing images, by performing dimensionality reduction processing on high-dimensional hyperspectral remote sensing images, combined with sample data to classify deep forests Model training, and then through the trained deep forest classification model to achieve the purpose of accurate and fast classification of hyperspectral remote sensing images; a classification device and storage device for hyperspectral remote sensing images, used to realize a hyperspectral remote sensing image Classification. The beneficial effects of the present invention are: by using the deep forest algorithm to classify hyperspectral remote sensing images, the classification accuracy and classification speed of the deep forest classification model are improved, and the method provided by the present invention involves few parameters, is easy to adjust, and greatly improves Classification effect and practicality of hyperspectral remote sensing images.

A method based on anemometer tower measurement data for determining the discarded wind electrical power of wind farms. On the basis of anemometer tower measurement data, the method takes into consideration the impact of such factors as terrain, topography, and wind turbine wake, calculates wind speed at wind turbine hub height, and obtains a theoretical wind turbine power according to a power curve, then adds up the theoretical power of all the wind turbines of a wind farm to obtain a total theoretical power thereof; finally, the total theoretical power of the wind farm and the actual power thereof are respectively integrated over time, to obtain a theoretical and an actual electrical power; subtracting one from the other yields the discarded wind electrical power of the wind farm, thus providing precise calculation of wind farm discarded wind electrical power. The present method is applicable to all types of wind farm and resolves problems of low accuracy in the calculation by the sample turbine method of discarded wind electrical power.