50results about How to "Reduce errors in judgment" patented technology

The invention discloses a coal equipment fault early warning system and method based on machine learning. A data signal acquisition unit acquires real-time data and acquires static data; a database acquires data from the data signal acquisition unit, data is organized according to a hierarchical structure of mine equipment to form an equipment hierarchical data model, and the equipment hierarchical data model is stored; a machine learning platform is connected with the data signal acquisition unit and/or the database for communication, and a machine learning basic algorithm and an operation environment can be provided; a machine learning algorithm module analyzes data and establishes an equipment fault early warning analysis model through the machine learning basic algorithm on the machinelearning platform and the operation environment, and the equipment fault early warning analysis model is used for equipment fault early warning and judgment; a fault early warning and judgment resultof the machine learning algorithm module is displayed through a result display unit. Early-stage abnormal conditions of the equipment can be found in time, and an early-warning message is released according to a detection result.

The invention discloses a voltage sag reason recognition method based on typical voltage sag waveform matching, and the method comprises the following steps: (A), collecting a historical monitoring waveform of a typical voltage sag historical event; (B), building a typical voltage sag event waveform library; (C), extracting a to-be-extracted voltage sag waveform; (D), carrying out the waveform matching and analysis of the wave band of the to-be-extracted voltage sag waveform and the typical voltage sag event waveform library; (E), recognizing the generation reason of a current voltage sag event waveform. The method carries out the more fundamental and accurate matching analysis of the voltage sag at the aspect of waveform, can quickly and accurately recognize the generation reason of the voltage sag event waveform, has a very high reference value for the prevention and treatment of the voltage sag, and is good in application prospect.

The invention provides a method for identifying a voltage sag reason. The method for identifying the voltage sag reason comprises the following steps of (1) using the discrete Fourier transform to calculate a spectral density function, converting the voltage in a time domain to the voltage in a frequency domain, (2) calculating the characteristic quantity related to identification of the voltage sag reason, (3) defining a sag phase number N1, judging whether swell and a sag combination N3 exist or not, and (4) identifying the voltage sag reason. The method for identifying the voltage sag reason is comprehensive and higher in adaptability to changes of voltage sags, and can be easily improved, verification is carried out on the basis of really-measured data of a power grid, and the calculation result has better engineering practicability.

The invention discloses a virtual DCS simulation system and a simulation method thereof. The virtual DCS simulation system comprises a DCS control system and a control virtual controller cluster. The DCS control system comprises a picture monitoring module and a configuration module, wherein the picture monitoring module is used for terminal operating and picture monitoring, and the configuration module can provide a configuration environment to achieve a simulation function. The control virtual control cluster is communicated with the DCS control system. The virtual DCS simulation system further comprises a simulation virtual controller cluster which can construct different physical equipment models and process models in the configuration environment. The virtual DCS simulation system further comprises a simulation platform, wherein the simulation platform can schedule and manage the control virtual controller cluster and the simulation virtual controller cluster in a unified mode. The virtual DCS simulation system can truly represent machine set characteristics, can bring visual training feel to trainees and can improve training effects.

The invention relates to a breakout prediction method for a slab continuous casting mold based on withdrawal resistance, belonging to the field of continuous casting link control during the metallurgical process and mainly being used for solving the problem of time lag in the existing breakout prediction method based on temperature. The breakout prediction method comprises the following steps: (1) collecting production data at continuous casting site, and calculating the withdrawal resistance; (2) de-noising a resistance signal, respectively extracting resistance features in the cases of normal and abnormal (breakout and casting powder) continuous casting processes, and then constructing feature vectors; (3) training a resistance signal recognition model by means of the feature vectors and a support vector machine (SVM); and (4) transferring a real-time withdrawal resistance signal into the recognition model so as to judge the production condition of the slab continuous casting mold. The breakout prediction method has the beneficial effects that the breakout situation is judged and predicted based on the withdrawal resistance so as to obviously advance the prediction time, thus being beneficial to preventing breakout accidents, improving the casting blank quality and optimizing continuous casting technological parameters.

The invention relates to a public opinion positive and negative classification method, and belongs to the technical field of information security. The method comprises the steps of collecting and preprocessing public opinion data in a target field, wherein the public opinion data comprises a plurality of public opinion articles; constructing a target domain emotion dictionary based on the BosonNLPemotion dictionary; obtaining emotion tendency scores of the public opinion articles through the target domain emotion dictionary, and judging positive and negative public opinion articles accordingto the emotion tendency scores; utilizing the public opinion data judged as the negative public opinion articles to construct a wrong score correction model; and performing positive and negative classification on the public opinion data by utilizing the error correction model.

The invention discloses a nomogram model for predicting the curative effect of tumor immunotherapy and an establishment method thereof. The method comprises the following steps: analyzing and screening patient mutant genes and clinical characteristic data to obtain factors with remarkable correlation with immunotherapy, and further constructing a corresponding nomogram model. The invention provides three nomogram models in total, is suitable for different non-small cell lung cancer patients, and assists in judging whether different patients can continue to receive treatment or select medicinechange; the model provided by the invention is simple and visual and easy to popularize and apply, can effectively help clinicians to carry out accurate individualized evaluation on Atezolizumab immunotherapy when non-small cell lung cancer patients are subjected to immunotherapy grouping and after treatment is started, brings better survival benefits to the patients, and has important value for effective application of non-small cell lung cancer immunotherapy.

A sensing home plate, structurally assembled from a home plate main body, and a sensing device disposed therein. The sensing device is assembled from a power supply circuit, a signal transmitting circuit, a fine tuning circuit, a signal receiving circuit, and an acousto-optic circuit. The signal transmitting circuit transmits signals, and forms a sensing area on the upper side of the home plate main body. When a baseball passes through or contacts the sensing area causing interference therewith, then a signal is received by the signal receiving circuit, and the acousto-optic circuit emits sound and light as a means to indicate such. The present invention can thus be used in the sport of baseball to accurately determine whether or not the ball pitched by the pitcher is a strike (good pitch) or ball (bad pitch), thereby assisting the chief umpire in making decisions, and reducing judgmental errors.

The invention provides a lighting control method and a lighting control system. An image acquisition module identifies whether a person is in a house or not. When a person is identified in the house,the state of the person is identified. When the person is in a sleep state, the lighting is automatically turned off, and when the person is in an active state, the lighting is kept on. When the imageacquisition module recognizes that no one is in the house, the Bluetooth module searches for and detects the Bluetooth equipment registered with the main control chip in the house. When detecting, the Bluetooth module judges that there is someone in the house and keeps the light on. When detecting is not possible, the Bluetooth module judges that there is no one in the house and automatically turns off the light. The application judges whether there is someone through the Bluetooth module more accurately, and reduces the inconvenience to the user caused by the error judgment of the image acquisition module and the automatic turning off of the light. At the same time, the image acquisition module and the Bluetooth module are used together, so that the system has the anti-theft monitoring function.

The invention discloses a steering testing device for a locomotive fan. The steering testing device comprises a wind speed sensor and a wind speed displayer which are connected with each other by a signal wire, wherein the wind speed sensor and the wind speed displayer are fixed on a shell; the wind speed displayer is positioned at the bottom of the shell; the shell is provided with a retractable rod; the wind speed sensor is mounted on the top end of the retractable rod; and the signal wire passes through the inside of the shell to connect the wind speed sensor and the wind speed displayer, so that a tester can conveniently and correctly judge the steering of the fan without approaching the fan, thereby reducing the appearance of errors and improving the dispatch quality of a locomotive. The wind speed sensor and the wind speed displayer can be different in models, and the retractable rods can be different in forms as well, so that the steering testing device has a wide application range.

The embodiment of the invention discloses a traffic signal identification method, a traffic signal indication device and a terminal, which are applied to the field of intelligent driving. According to the traffic signal indicating device provided by the embodiment of the invention, besides a visible light signal indicating system in an existing traffic light mode, the near-infrared light source is additionally arranged, and the vehicle-mounted terminal is based on the collected near-infrared light signal; and traffic operation information indicated by the traffic signal indicating device is determined. As the penetrating power of the near-infrared light signal is higher than that of the visible light signal, the recognition accuracy of the traffic signal can be improved.

The invention belongs to the technical field of face recognition, and particularly discloses an intelligent sign-in system integrating face recognition and data analysis. The sign-in system comprisessign-in equipment, a face recognition end and a mental degree evaluation end. The face recognition end is used for recognizing, extracting and comparing face information. The mental degree evaluationend is used for collecting and analyzing a face image and evaluating the mental degree of the sign-in person. The sign-in equipment is used for collecting the face information. The intelligent sign-insystem is scientific and reasonable and safe and convenient use. The mental degree evaluation end is arranged, two modes of data analysis and machine learning can be used for evaluating the mental degree of the sign-in person, the evaluation result can be sent to the client to be displayed, the manager can arrange days off for the employees with a poor mental state according to the displayed mental degree to reduce the error probability in work, the employees can feel warmth of the enterprise and the work efficiency can be effectively improved.

The invention discloses an athletic starting block capable of adjusting the starting height and a method thereof, belongs to the technical field of athletics, and aims to solve the problems as follows: an existing starting block is troublesome in adjusting process, the starting height cannot be adjusted in real time according to needs, and the overall adaptability cannot be adjusted at will according to needs; and secondly, the pedal of the existing starting block is not subjected to anti-skid treatment, so that the pedal easily falls down during starting, and the running performance is influenced. An inclined pedal can be effectively adjusted and positioned in a directional mode, and dislocation of the inclined pedal is prevented. Side-by-side tooth grooves are formed in the front end of the inclined pedal, friction force of the whole pedal can be greatly increased, the starting stability is improved, falling during starting is prevented, and the starting quality is improved. An abutting block drives the inclined pedal to adjust the inclination angle of the inclined pedal when moving, and thus, the starting height can be adjusted. The starting height of the inclined pedal is adjusted to meet the starting requirements of different matches according to different requirements, and the deployability is high. The starting height can be adjusted according to different training, and the training efficiency is improved.

The invention provides a sending end equalization optimization method and device and electronic equipment, belongs to the technical field of computer networks, and solves the problems that people need to consume a large amount of time and energy to manually adjust parameters and misjudgment is possibly caused by experience insufficiency in the prior art. The method comprises the steps of initializing an equalization parameter population of a sending end, wherein the equilibrium parameter population at least comprises a key equilibrium parameter combination of one sending end, and the key equilibrium parameter combination at least comprises one key equilibrium parameter; dividing all the key equalization parameter combinations into a plurality of preset blocks; screening out a target block from the plurality of preset blocks by using a gene algorithm; determining an optimal equilibrium parameter combination from the target block by using a hill climbing algorithm, and determining an optimal equilibrium value based on the optimal equilibrium parameter combination. According to the method, the judgment errors of personnel, especially inexperienced new people, are greatly reduced, and the fine-tune time of the personnel is greatly reduced.

The invention provides a catheter tail end attitude estimation method and system and a medium, and the method comprises the steps of obtaining the prior shape of the catheter tail end in a three-dimensional scanning or manual surveying and mapping mode before intubation, cutting a two-dimensional projection drawing of the catheter tail end in an XCA image in the intubation process, and recording the direction information of the two sides of a bent section of the catheter tail end; fusing the prior shape of the tail end of the catheter and the two-dimensional projection drawing information at the current moment in the front fusion stage, and fusing the direction information of the electromagnetic sensor at the current moment into the middle layer of the deep learning network and fusing with the middle layer features in the rear fusion stage; and inputting the pre-fusion result into an encoder of the deep learning network, inputting the post-fusion result into a decoder of the deep learning network, and finally outputting attitude estimation in a point sequence format by the deep learning network. According to the invention, the deep neural network is firstly used for estimating the catheter tail end posture, and the method has the advantages of being high in robustness, high in calculation efficiency, accurate in estimation and the like.