723results about How to "Clear thinking" patented technology

The present invention belongs to the technical field of concrete construction and particularly relates to a design method for a mixing proportion of a self-compacting concrete mixture. The method comprises the steps of: determining a performance design target; determining mixing strength and a water-binder ratio; determining densest gradation (or a sand rate) of aggregates and void ratios of different packing states; determining the usage amount of binding materials and mixing water; determining the usage amount of the aggregates; determining the usage amount of additives; determining the actual adding amount of the mixing water; and the like. According to the method provided by the present invention, aimed at blindness of material selection during self-compacting concrete mixing proportion design in the prior art, based on a fundamental theory of particulate dense packing and a composition rule of reaching a densest packing state of solid particulate materials with different particulate diameters, a mixing proportion of different materials is determined by an actually measured densest packing state, and relationships are established between the material mixing proportions corresponding to different densest packing states and expected self-compacting performance grades of the mixture, so that the design method used for the mixing proportion of the self-compacting high-performance concrete mixture and based on a use performance as a target is constructed.

The invention relates to an ultra-high voltage direct current transmission line lightning stroke interference recognition method based on voltage relevancy and wavelet transformation transient state energy distribution characteristics, and belongs to the technical field of power system relay protection. When voltage of an ultra-high voltage direct current transmission line fluctuates, positive voltage and negative voltage of the initial 5ms are monitored; the relevancy between voltage of each electrode and voltage of each axis is calculated, and the lightning stroke interference is recognized according to the relevancy; the recognition of fault classifications is shifted to if the interference is caused by a fault; an additional component of line modal voltage is calculated, and seven dimension decomposition is performed on the additional component by utilizing a db3 wavelet; and the ratio of high frequency energy to low frequency energy is calculated, fault categories are distinguished, and fault protection is performed. According to the method, the signal acquisition time is short, the sample acquisition scale is small, and the voltage relevancy theory is used for recognizing lightning stroke interference and fault, so that the thought is clear, and the accuracy is high; the transient state energy distribution characteristics obtained through wavelet transformation are used for recognizing lightning stroke fault and short circuit fault, and the effect is obvious; and the method is easy to implement and is not easily influenced by system parameters and transition resistance.

The present invention discloses parameter regulating method of non-linear electricity system stabilizer based on frequency domain test and pertains to the technical field of stability control of electric power system. The present invention is characterized in that: test the non-frequency-compensation linear property curves and test the phase angle under different frequency points in the low frequency vibrating range of 0.2 to 2.0Hz from the output point of the stabilizer to the voltage of the generator after the working condition of tested machine unit is fixed; then calculate the lead and lag angle a 1 of the linear part of the output signals of the stabilizer under different frequencies, added with phase angle that has corresponding frequency property, and then decide whether surpass the lead and lag range relative with Delta Omega axle prescribed by national standard; if not surpass, the feedback factor k1, k2 and k3 of the linear part can be calculated out; then the damping coefficient C1 and the gain coefficient C2 can be decided with critical amplification times method and load phase step method. The present invention is easy to be comprehended by commissioning operators and is easy to be populated.

The invention discloses pervious concrete of recycled aggregate and a preparation method thereof, and belongs to the technical field of pervious concrete of recycled aggregate. The pervious concrete of the recycled aggregate comprises the following components by mass percent: 71.77-73.16% of coarse aggregate, 18.88-19.04% of cement, 6.58-8.14% of water and 1.21-1.22% of silica fume, wherein the recycled aggregate and the natural aggregate accord with the stone utilization standard II, the mixing amount of the recycled aggregate can be customized according to the actual performance index, the cement is 42.5-level ordinary portland cement, and the mixing amount of the silica fume is 6% of the dosage of a binding material. The preparation method comprises the following steps: agitating, vibrating, molding and floating. The mix proportion is designed by adopting a volume process, and the concrete is prepared according to the scheduled preparation process. By adopting the pervious concrete of the recycled aggregate, on the one hand, the problem of recycling of the waste concrete can be solved, on the other hand, underground water can be effectively supplied, collected and purified, and water resources are repeatedly utilized, so as to relieve the 'heat island effect'.

The invention discloses a micro-grid short-term load forecasting method based on the EMD-KELM-EKF, and belongs to the technical field of intelligent power grids. The method includes the steps of firstly, preprocessing micro-grid history load data, enabling the preprocessed micro-grid history load data to serve as a training sample, and extracting the load attribute of the training sample; secondly, determining a combined forecasting model based on the EMD-KELM-EKF and parameter initial values of the combined forecasting model; thirdly, optimally selecting parameters of the combined forecasting model through a particle swarm optimization algorithm; fourthly, substituting a test sample into the combined forecasting model based on the EMD-KELM-EKF with determined parameters to obtain a forecasting result. The method is clear in physical concept and thought, and provides the theoretical foundation for micro-grid short-term load forecasting, and calculation is easy and convenient; according to the method, by means of the implementation pattern of combining offline parameter optimization with online forecasting, the forecasting result is good in stability, and optimal dispatching can be easily conducted on micro-grids.

The invention discloses a laser radar two-step calibration method based on a calibration field. The method comprises steps of establishing a calibration field; carrying out calibration; preparing angle calibration; networking a theodolite; scanning a laser radar; measuring the theodolite; establishing an angle correction equation; calculating an angle correction coefficient; establishing a base line field; measuring the laser radar; measuring the theodolite; establishing a distance correction equation; and calculating a distance correction coefficient. According to the invention, rapid calibration of the laser radar is achieved without taking a precisely designed and processed calibration device as a calibration standard; a problem of too high requirements on a test place, the calibration field and the calibration device in the normal method is solved; through cooperation of a rotation bench, the whole test process is achieved, the test difficulty level is reduced and a difficulty in ensuring precision of full field-of-view angle correction in the normal method is solved; and through the test of the calibration field and the base line field step by step, effects of correcting separation distances in a calibration model are achieved and disadvantages in mutual coupling of distance parameters and angle parameters in the normal method are overcome.

The invention provides an opportunistic routing method capable of efficiently utilizing energy, for a mobile wireless sensor network, which solves the problems of node holes, short network lifetime and the like of the mobile wireless sensor network, so as to prolong the lifetime of the mobile wireless sensor network. According to the method, a candidate forwarding node set is determined, the priority of the candidate forwarding node set is optimized by the comprehensive measurement strategies of energy estimation, load estimation, movement influence factors and the like, and in combination with the mechanisms of geographic positions, energy sensing, load balancing and the like, so as to finish the data package forwarding of the mobile wireless sensor network. The method disclosed by the invention is capable of effectively alleviating the problems of node holes, short network lifetime and the like of the mobile wireless sensor network.

The invention relates to a ship track control method based on an improved LOS guidance algorithm. The method comprises the following steps: 1, obtaining the real-time navigation information of a ship by using the navigation module of the ship when the ship sails along a certain straight track segment in a path generated by set track points; 2, operating the improved LOS guidance algorithm to calculate a desired course when a track deviation between an actual track and a planned path exceeds an acceptable deviation d' or keeping an original planned track of the ship to sail if not; and 3, repeating the step 2 until the track deviation is within the acceptable deviation. The method applies the improved LOS guidance algorithm to the field of ship track tracking control, is clear in thinking and simple in algorithm, can control the ship to track the set ship course, is smooth in rudder angle change, and can avoid frequent steering of the ship.

The invention provides a light field imaging and printing device and a micro lens thin film with three-dimensional floating images, wherein the thin film is prepared by the light field imaging and printing device. The scanning operation on a lens surface is conducted by a scanning galvanometer system, and four-dimensional light field data can be printed in combination with the variation of space three-dimensional coordinates. The light field imaging and printing device comprises a light source, a scanning galvanometer system, a lens, a diffuser, a converging lens assembly and a micro lens recording material. Light rays sequentially pass through the scanning galvanometer system, the lens, the diffuser and the converging lens assembly, and then instantaneous and dynamic focus spots are outputted. In this way, a volumetric-pixel type radiation light field is obtained. After that, the information of the instantaneous volumetric-pixel type radiation light field is recorded by the micro lens recording material. By means of the device and the thin film, more realistic three-dimensional dynamic images can be formed. Through controlling different variable values, floating images of multiple different characteristics are formed. Therefore, the multi-view imaging effect of varying images can be realized.

The invention provides a distributed online real-time processing method and system for multi-source and heterogeneous flow-state big data. The method specifically comprises the steps of crawling webpage data of each source by utilizing a distributed crawler duplicate removal algorithm; pre-processing the crawled page, constructing a corresponding tree by utilizing a visual page segmentation algorithm, pruning noise nodes according to a visual rule, classifying multi-layer pages, determining predicates under different types of pages according to different characteristics, and inferring data recording block nodes and data attribute nodes through the rule; Distributing the preprocessed data source by using a distributed message system, providing a data stream, and describing the state of a data node in the data stream to form state information; Selective storage operation is carried out on data streams by utilizing a Hadoop distributed file system based on K- And detecting the processed data by using a means text clustering method, determining a text similar to a predetermined sensitive information text, and screening out the sensitive information.

The invention disclsoes a CTCS-3 level train control system integrated test method. According to the method, starting from key and necessity test items of an integration test, the integration test time and a driving route are combined, and CTCT-3 level train control system integration test is arranged as a whole.The method is composed of five steps, including a C2 level 90km/h pull-on test; a C3 level 90km/h pull-on test; a mobile authorization test; C3 level inter-level switching and a station track shunting test; and a manual route canceling and station side area speed limiting function test; the overall function of a C3-level system can be verified in a short time, functions of C3-level subsystems and the interface configuration function between the C3-level subsystems can be verified, the communication quality of train-ground communication can be verified, and a basis is provided for smooth joint debugging and joint testing of a signal system; meanwhile, each step is clear in thought, simple and clear in content, easy to understand and high in operability. The method is advantaged in that the universal sample plate for the C3-level integration test is provided, a solid foundation is laid for smooth development of subsequent joint debugging and joint test, a test method is provided for the integration test of the C3-level high-speed railway, and the high-quality effect of the integration test is guaranteed.

The invention discloses an intelligent operation and maintenance framework design method for a power grid dispatching control system, which includes the following steps: 1) omni-directional monitoringdata acquisition, 2) monitor data storage, 3) real-time monitoring alarm, 4) fault diagnosis and location, and 5) multi-dimensional visual display. The application can monitor the operation status ofthe whole system in real-time, evaluate the health status of the software and hardware such as the host, storage, application and the like in an effective, accurate and timely manner, and provide effective information for system optimization and fault discovery, and improve the operation and maintenance efficiency of the system.

The invention discloses an urban water disaster risk prediction method based on an RBF (radial basis function) neural network-cloud model. The method includes (1) determining evaluation factors, levels and the indicator range under corresponding levels; (2) determining an expectation Ex and an entropy En of the cloud model; (3) determining the weight of each evaluation factor according to measured values of the evaluation factors and the indicator range of each level; (4) training the RBF neural network, finishing model establishment for the RBF neural network, inputting the measured values of the evaluation factors of the cloud model to the trained RBF neural network to perform simulated prediction, and obtaining a prediction value of each evaluation factor; and (5) substituting the prediction value of each evaluation factor to the integrated cloud model to allow the integrated cloud model to calculate corresponding certainty degree of the prediction value of each evaluation factor belonging to each risk level and multiply the corresponding weight to obtain integrated risk level distribution. The urban water disaster risk prediction method is visualized and reliable and strong in operability, and accuracy of prediction is improved.

The invention discloses a method for tuning and optimizing parameters of a dynamic absorber based on a machining process, which comprises the following steps: firstly, establishing a mathematical model of two degrees of freedom of a vibration system, and then solving the amplitude of the steady-state solution of a dynamics differential equation; taking the relative displacements of a main structure and the dynamic absorber in the viscous damping energy consumption formula; drawing an energy curve, and then making a statistics of the relation between a single factor vibration and the energy peak value; according to the statistics data, fitting the optimal frequency ratio and the optimal damping ratio corresponding to each mass ratio; according to the frequency range of an external disturbance signal, performing integration on the energy that the dynamic absorber can absorb; then, drawing an energy histogram and finding the maximum value so as to obtain the optimal mass ratio; finally, obtaining all optimal parameters of the dynamic absorber. With adoption of the method disclosed by the invention, the vibration caused by external disturbance can be restrained, so that the response speed can be improved by about 50%.

The invention discloses a hierarchical structure-based neural-network machine translation model, and relates to natural language processing based on deep learning. A word alignment tool GIZA++ is used to carry out word alignment on parallel training sentence pairs, and then source language sentences are divided into clauses of monotonous translation according to punctuation and word alignment information; the above-mentioned obtained clause data are used to train a clause classifier; hierarchical structure modeling is carried out on the source language sentences of the parallel sentence pairs; and hierarchical structure decoding is carried out on target language sentences of the parallel sentence pairs. The sentences are divided into the clauses of monotonous translation, and then word-clauses-sentence hierarchical-modeling, attention mechanisms and decoding are carried out: and a bottom-layer recurrent neural network (GRU) encodes semantic representations of the clauses, an upper-layer recurrent neural network encodes information of the sentences, the bottom-layer attention mechanism is devoted to word-level alignment inside the clauses, and the upper-layer attention mechanism is devoted to clause-level alignment.

The invention relates to a cascading-type STATCOM (Static Synchronous Compensator) DC (Direct Current) side capacitor voltage balance control method. In the control method, the direct current control and the indirect current control are combined. The method comprises the following steps: generating an active current reference value; selecting a special bridge; generating an STATCOM output current reference value; carrying out indirect current control of the special bridge; and carrying out direct current control of a non-special bridge. The control method provided by the invention is used for controlling the cascading-type STATCOM DC side capacitor voltage balance. The method not only ensures the balance of a DC capacitor voltage between H bridges and improves the response speed of a control link and optimizes the system performance, but also does not require an additional hardware circuit and cannot ensure the hardware cost of a STATCOM device to be increased.

Aiming at solving the problems that flow velocity and impact of any point in any section of mud-flow cannot be measured in the prior art, the invention first provides a method for measuring viscous mud-flow internal flow velocity and impact force spatial distributions. The method for measuring flow velocity is based on the viscous mud-flow body Bingham model and the scour-and-fill balance principle; first, setting any section I-I, determining values such as h0, h(x), xa, xd, xb, xc of the section through the trial method; then calculating the flow depth of a mud-flow calculation infinitesimal surface at any point on the section according to formula 1; and finally substituting h into the Bingham-model-based viscous mud-flow flow velocity distribution formula to obtain flow velocity of the point. By means of the method for measuring flow velocity at any position in any section of mud-flow, the method for measuring spatial distribution of viscous mud-flow impact force is further obtained. The calculation result of the method has high accuracy, which can be applied in real design for viscous mud-flow prevention and control engineering. The measurement method is clear in thinking and simple in calculating, which is an improvement of the present prevention and control engineering design method which is based on uniform distribution.

The invention relates to an infrared focal plane non-uniformity fingerprint extraction and image correction method which belongs to the technical field of infrared imaging. A production process decides, each infrared focal plane is provided with a relatively stable non-uniformity mode and a rule thereof which varies with temperature, and the two are collectively called fingerprints. The invention provides the definitions of non-conformity fingerprints, analyzes the characteristics of the focal plane non-uniformity in a frequency domain by wavelet decomposition, extracts the non-uniformity fingerprints which belong to the focal plane, and then stores the fingerprints in a storage unit of a detector. When in correction, with the environment temperature as the input parameter, the method can carry out non-uniformity correction under the constraint of the fingerprints after reading the non-uniformity fingerprints at the environment temperature. Compared with convention methods, the method does not need to obtain a background frame with a uniform baffle in every time of correction and has simple correction idea, thereby simplifying the correction device and the correction process, and the image non-uniformity after correction is significantly improved.

The invention discloses a disturbance suppression method based on a compound disturbance observer, and aims at the problem that system requirements can not be stably controlled at high accuracy sincethe low-frequency and middle-frequency disturbance suppression capability of a control system stable platform is poor. By use of the method, on the basis of multi-closed-loop control, a traditional disturbance observer design is improved, the compound disturbance observer structure is put forward, and a design method for an inner and outer ring feedforward compensation controller in the compound disturbance observer is given. By use of the method, an outer ring in the compound disturbance observer is utilized for estimation, inverse feedforward of a middle-frequency band and external disturbance, and an inner ring is used for estimation, inverse feedforward of a low-frequency band and external disturbance. The inner ring and the outer ring of the compound disturbance observer simultaneously act, finally, the purposes of simultaneously improving the disturbance suppression capability of the low frequency band and the middle-frequency band of the system and improving and controlling system stability accuracy are achieved.

The invention discloses a method and a device for positioning an iris image. The method comprises the following steps: carrying out rough pupil positioning on the iris image; initializing coordinates of a plurality of key points based on rough positioning; extracting SIFT feature vectors of the key points; using an SDM algorithm to map the SIFT feature vectors of the key points into a coordinate increment; using the coordinate increment to update coordinate values of the key points; and using a robust regression method to carry out curve fitting on the key points to obtain parameters of an upper eyelid parabola, a lower eyelid parabola and an iris excircle. Compared with the prior art, the method for positioning the iris image disclosed by the invention is simple, convenient, high in speed, strong in anti-interference ability, high in precision, high in generalization performance and good in robustness.