32results about How to "Dealing with uncertainty" patented technology

A load flow calculation method of a distributed power supply connection power grid comprises the steps that S1. initial data of an electric power system are read; S2. sampling frequency N and the dimensions s of input random variables are determined; S3. an s * N order sampling matrix is generated; S4. sampling frequency is initialized, namely n is equal to 1; S5. whether n is larger than the sampling frequency N is judged, and yes, the probability statistics results of the variables are directly output; otherwise, S6 is carried out; S6. a wind power and photovoltaic power generation output model is determined, and a load random model is determined; S7. a load flow calculation model is determined; S8. an optimized economic model is determined; S9. load flow calculation is carried out; S10. data such as voltage, branch power and power generation cost of a <n>th node group are determined; and S11. a next round of load flow calculation is carried out, t is equal to t + 1, and S5 is carried out. The probability distribution of the output random variables can be estimated well, the uncertainty problem in an electricity market can be well solved, debugging manpower and material resources are saved, and production cost is lowered.

The invention relates to the field of visual servo systems, in particular to a PD-SMC control method of a visual servo system based on an eye-on-hand structure so as to solve the problems of uncertainty and large computation amount existing in an existing visual servo system control method. The method comprises the steps of (1) setting an expected image and collecting feature points of the expected image as an expected visual feature set; (2) conducting projection transformation on a target expected image with the current state of the relative poses of a camera and a target body to obtain a current image of the camera; (3) adopting the PD-SMC method for a visual servo controller part and adopting a proportional control method for a joint controller part; (4) extracting a visual feature setfrom the current image, computing a jacobian matrix of the image and then computing the controlled quantity u of the camera in a Cartesian space; and (5) computing the pose of the target body in a camera coordinate system, and repeating the operations with the pose as the current pose until the error is zero. The PD-SMC control method can be applied to target grabbing equipment.

The invention discloses a GIS fault diagnosis and reliability analysis method based on fuzzy Petri. According to the method, a GIS fault modeling and inference method of a fuzzy Petri net is put forward on the basis of the randomness and diversity of GIS faults and the fuzziness of fault symptoms, the theory of the Petri net and the fuzzy inference rules are organically combined together, and the high-efficiency and accurate parallel inference capacity is achieved. Through the combination of lots of statistics fault cases, the fault diagnosis model of a GIS based on the FPN is established. When the system does not fail, matrix forward reasoning is conducted on the fault state of the GIS by online monitoring and predicting the possible fault symptoms, the credibility of all places can be rapidly and accurately calculated, and the importance degree of the initial place is analyzed; fault reasons are found through backward fuzzy inference rules under the known fault phenomenon.

The invention discloses a mechanical equipment fault diagnosis method based on deep learning. The method specifically comprises the following steps of S1, carrying out the data collection and preprocessing of a main data source and a secondary data source of mechanical equipment, and obtaining a data set; S2, a five-fold cross validation method being adopted to divide the data set into a trainingset, a validation set and a test set; and S3, establishing a fault diagnosis model based on the CNN and the BD-LSTM, inputting the training set into the fault diagnosis model, extracting hidden features, performing training, and outputting a diagnosis result. According to the method, BD-LSTM is adopted to perform smooth tracking and result prediction, and uncertainty caused by operation and environmental interference is processed, sensor monitoring data adopts CNN and BD-LSTM to extract hidden features in parallel, output of two irrelevant paths can influence prediction, and each parameter inthe network can be corrected according to predicted errors.

The invention discloses a multi-layered elite character method for mining special medical history diagnosis rules of the traditional Chinese medicine. The multi-layered elite character method comprises the following steps of designing the concentration selection probability of elite character sub-groups; distributing attributes of medical histories of the traditional Chinese medicine into 'common-elite' character groups with different classes; performing pretreatment on related attributes and correlative dependence attributes in the special medical histories of the traditional Chinese medicine; establishing a dynamic balancing strategy based on multi-layered elite characters; selecting an elite subset vector with the global search and local refining maximum optimal capability from the multi-layered elite characters; and establishing a maximum elite optimizing array to quickly mine the special medical history diagnosis rules of the traditional Chinese medicine. By the multi-layered elite character method for mining the special medical history diagnosis rules of the traditional Chinese medicine, problems of attribute fuzzy, attribute correlation and correlative dependence of special electronic medical records of the traditional Chinese medicine can be solved well, the diagnosis rule mining efficiency is improved effectively, and the robustness and the practicality are high.

A multi-scene analysis-based reactive power optimization method for a power distribution network of a wind power generation set comprises the steps of constructing a reactive power optimization mathematical model for the power distribution network of the wind power generation set, wherein the reactive power optimization mathematical model comprises a reactive power optimization target function and a reactive power optimization constraint condition; determining the reactive power optimization target function by employing a multi-scene analysis method according to output charge of the wind power generation set and load fluctuation; and solving the reactive power optimization mathematical model by a particle swarm optimization method. According to the method, the output change of the wind power generation set and random fluctuation of a load are fully considered, the output of the wind power generation set and the load are divided into a plurality of sections to form a plurality of scenes by constructing the reactive power optimization model for the power distribution network of the wind power generation set and employing the scene analysis method, the minimum expected value of active power network loss in the scenes is used as an optimization target, and then reactive power optimization is performed by the particle swarm optimization method so that the method is suitably used for processing reactive power optimization of the power distribution network of the wind power generation set.

The invention discloses an unexpected event emergency capacity assessing method based on interval binary semantics. The unexpected event emergency capacity assessing method comprises the following steps: firstly, building an emergency capacity assessing index system, and building an interval binary semantics preference relationship relevant to an assessing index through pairwise comparison; then,inspecting and correcting the consistency of the interval binary semantics preference relationship by an iterating algorithm; exporting assessing index weights and calculating a sub-index global weight; taking subjective and objective weights of a decision maker into consideration, and calculating a group sub-index global weight by using an interval binary semantics aggregation operator; acquiringan interval binary semantics assessing value of a scheme to be assessed on a sub-index; taking the subjective and objective weights of the decision maker into consideration, and calculating a group scheme assessing value by using the interval binary semantics aggregation operator; and integrating interval binary semantics interval set aggregation operators to obtain comprehensive assessing valuesof schemes for comparison and ranking. Through adoption of the unexpected event emergency capacity assessing method, information loss can be avoided, and the complexity, uncertainty and thought fuzziness of the decision maker during emergency capacity assessing can be well processed.

The invention discloses a multi-point temperature intelligent detection system which is composed of an environmental parameter acquisition platform and a multi-point temperature big data prediction subsystem, the environmental parameter acquisition platform realizes environmental parameter detection, and the multi-point temperature big data prediction subsystem realizes environmental parameter processing and prediction. The accuracy and the reliability of environment temperature detection are improved; the problem that industrial and agricultural economic benefits and environment temperature management are greatly influenced due to the fact that an existing environment temperature parameter detection system does not influence temperature monitoring accuracy and reliability according to nonlinearity, large lag, large environment and the like of environment temperature parameter changes and does not dynamically predict environment temperature parameters is effectively solved.

The invention discloses an intelligent building settlement detection system which is composed of a building settlement parameter acquisition platform based on a wireless sensor network and a buildingsettlement intelligent early warning system, and the building settlement parameter acquisition platform based on the wireless sensor network realizes detection and management of building settlement parameters. The system effectively solves the problems that existing building settlement has no influence on the settlement of the whole building according to nonlinearity, large lag, complex settlementchange and the like of settlement amount change of each detection point of the building; accurate detection, prediction and early warning are not carried out on building settlement, so that early warning and management of the building settlement amount are greatly influenced.

The invention provides a cowhouse environment temperature detection system based on a wireless sensor network which consists of a cowhouse environment parameter collection platform based on a wireless sensor network and a cowhouse environment multi-point temperature fusion model. The cowhouse environment parameter collection platform based on a wireless sensor network is used for cowhouse environment temperature detection, regulation and monitoring. The cowhouse environment multi-point temperature fusion model achieves cowhouse environment multi-point temperature fusion based on a similarity fusion weight, a distance fusion weight and a game theory combination weight calculated from a similarity matrix and a distance matrix of temperature Vague values of temperature sensors of multiple detection points in a cowhouse environment, so that the accuracy, reliability and robustness of cowhouse environment temperature fusion are improved. The system effectively solves the problems that a conventional cowhouse monitoring system does not detect the temperature of a cowhouse environment according to the characteristics such as nonlinearity and large lag of temperature changes of cow environment, large area of cowhouses and complicated temperature changes, so that cowhouse environment temperature detection is greatly influenced.

The invention discloses an intelligent bridge settlement detection system. The system is characterized in that the system is composed of a bridge settlement parameter detection platform based on a wireless sensor network and a bridge settlement early warning system. The bridge settlement parameter detection platform based on the wireless sensor network detects and predicts bridge settlement parameters. The bridge settlement early warning system is composed of a bridge settlement interval number neural network model, a bridge settlement prediction model and an interval number bridge settlementgrade classifier. According to the invention, the problem in the prior art that bridge settleability parameters are monitored only by equipment is effectively solved; only the bridge settlement parameters are obtained, the bridge settlement parameters are not processed according to the characteristics of nonlinearity, large lag, complex change and the like of the bridge settlement parameters, andthe problem that the service condition of a bridge is effectively predicted according to the influence of bridge settlement on bridge collapse deformation is solved.

The invention relates to the field of automatic control, and provides a fractional order adaptive control method and device for a time-delay multi-flexible swing arm system, and the method comprises the steps that a fractional order nonlinear model for a time-delay multi-flexible swing arm system is constructed; a communication network of the time-delay multi-flexible swing arm system is constructed through a directed topological graph; each order of tracking error of each flexible swing arm in the communication network is introduced; through the fractional order nonlinear model and the tracking error, a self-adaptive controller is obtained through calculation; and the motion of each flexible swing arm in the time-delay multi-flexible swing arm system is controlled through the self-adaptive controller. According to the method, by constructing the fractional order nonlinear model, the accuracy and complexity of the model can be comprehensively considered, and the time-delay multi-flexible swing arm system can be better described; a self-adaptive control method is provided, and uncertain factors in the time-delay multi-flexible swing arm system are processed; and an error barrier function and a radial basis function neural network are introduced to compensate a time lag factor in the system, so that accurate tracking of a plurality of flexible swing arms on a reference signal is realized.

The invention discloses a method for quickly solving sparse codes by using a hybrid sparse neural network. The method comprises the steps of S1 adopting a rough estimation neural network, the hybrid sparse neural network and a sampling module to jointly serve as a network solving model; S2 inputting the input data into a coarse estimation neural network for prediction to obtain an initial prediction result of sparse coding; S3 inputting the initial prediction result into the hybrid sparse neural network for prediction to obtain a probability distribution parameter of sparse coding; and S4 inputting the probability distribution parameter of sparse coding into a sampling module to sample a probability distribution result, and obtaining sparse coding according to the sampling result. According to the method, because probability distribution prediction and sparse coding solving modes are combined, compared with an existing sparse coding solving method, probability distribution of prediction coding can better describe coding, and sparse coding can be accurately solved.

The invention relates to the technical field of power distribution networks, in particular to a wind storage combined optimization configuration method based on extension distance K-mean clustering. The method comprises the following steps: combining an extension distance theory with a K-mean clustering algorithm to perform multi-scene analysis on wind storage combined optimization configuration;establishing a wind storage combined system optimal configuration model based on multi-scene analysis, source network load collaborative optimization and differentiated demand response; and introducing the idea of a sine function and a parallel computing technology into a differential evolution algorithm, and constructing a parallel multi-objective sine differential evolution algorithm to solve the wind storage combined system optimal configuration model. The extension distance K-mean clustering algorithm provided by the invention can effectively improve the accuracy and balance of the clustering result; a scene generated on the basis of an extension distance K-mean clustering multi-scene analysis method can effectively process uncertainty of distributed wind power output and load requirements; according to the method, the PMOSDE algorithm is adopted to solve the model, and the optimization speed and the optimization depth are effectively improved through the PMOSDE algorithm.

The invention discloses a big data detection system for livestock and poultry activity information. The big data detection system is composed of a livestock and poultry sign parameter acquisition and intelligent prediction platform based on a cloud platform and a livestock and poultry activity big data prediction subsystem. The livestock and poultry physical sign parameter acquisition and intelligent prediction platform based on the cloud platform is used for detecting and processing temperature and activity information parameters of livestock and poultry physical signs; the livestock and poultry activity big data prediction subsystem is used for predicting the activity state of livestock and poultry and providing data and early warning for preventing livestock and poultry diseases; the invention aims to provide the big data detection system for the livestock and poultry activity information, and the system monitors the livestock and poultry body temperature and activity information in real time, so that data and early warning are provided for preventing livestock and poultry diseases.

The invention discloses an environment big data Internet of Things intelligent detection system, which comprises an environment parameter acquisition platform and a formaldehyde big data intelligent prediction subsystem, realizes accurate detection and grade classification of formaldehyde concentration, and improves the reliability and accuracy of formaldehyde concentration detection. The system effectively solves the problems that an existing environment parameter detection system has no influence on the accuracy and reliability of measured environment parameters according to complex changes such as large environment area, nonlinearity of environment parameter changes, large lag and the like, and does not accurately detect and predict the environment parameters, so that the monitoring and management of the environment parameters are greatly influenced.

The invention discloses a livestock and poultry health sign big data Internet of Things detection system which is characterized in that the detection system comprises a parameter acquisition and control platform and a livestock and poultry body temperature big data intelligent prediction subsystem, and accurate detection and prediction of the measured livestock and poultry body temperature are achieved; the system effectively solves the problems that an existing livestock and poultry sign parameter detection system does not accurately detect and predict livestock and poultry sign parameters according to the influence on the livestock and poultry sign parameters due to the fact that the livestock and poultry environment area is large, and the livestock and poultry environment parameters and complex changes such as nonlinearity and large lag of livestock and poultry sign parameter changes are complex; the livestock and poultry health and the livestock and poultry management are greatly influenced.

The invention discloses an action scheme optimization method and device in a simulation training system, aiming at training requirements, activities in training are divided into action units in stages, the action units are adopted to construct a desired action scheme, and for each action scheme, the action scheme is optimized. Constructing an efficiency model corresponding to the action scheme by taking each action unit and the action result in the action scheme as an efficiency unit, mapping the efficiency model into a Bayesian network, performing parameter learning on the Bayesian network, taking parameters corresponding to Bayesian network nodes obtained by learning as probabilities of the efficiency units, and calculating the efficiency of the action scheme according to the probabilities of the efficiency units. And selecting the action scheme meeting the condition or adjusting the action scheme. According to the method, the uncertainty in the action scheme and the complexity of the dependency relationship between the actions are well processed, and optimization and optimization of the scheme are realized through multi-dimensional evaluation, so that reasoning decision is supported.

The invention discloses a multi-dimensional uncertain non-uniform medium target electromagnetic characteristic extraction method based on a volume-surface integral equation, and the method comprises the following steps: firstly, building a model for a target through employing a non-rational B-spline technology and ANSYS software, and further controlling the tiny deformation of the target through employing several control points as shape random variables; the change of the dielectric constant of the medium can be controlled through the random variable of the dielectric constant, then the random variables of the shape and the dielectric constant are introduced into a volume-surface integral equation through a disturbance method, and finally each disturbance current is iteratively solved by sampling the variable quantity of the random variable for multiple times; and calculating the radar cross section of the target model after the appearance or the dielectric constant is slightly changed, and the statistical mean and variance of all RCS responses. According to the method, the influence caused by tiny deformation of the appearance of the target or tiny jitter of the dielectric constant can be considered.

The invention relates to a spacecraft fault detection method based on Riemannian measurement, and belongs to the technical field of fault diagnosis. The method comprises the steps of 1, collecting N groups of off-line process data {Y1,..., YN} under the fault-free condition, and establishing a positive definite matrix rho (n) = [P1,..., PN]; 2, calculating a Riemannian center Pg of the positive definite matrix [P1,..., PN] through an iterative solution method; 3, sequentially calculating performance indexes Ji corresponding to the offline process data, and calculating a threshold value Jth through a threshold value setting algorithm; and 4, obtaining an online sample value, calculating a performance index J of the online sample value, comparing J with the threshold value Jth obtained in the step 3, if J is greater than or equal to Jth, giving a fault alarm, and if J is less than Jth, determining that no fault exists. According to the method, the fault is detected based on the batch data matrix, the matrix covers the mean value, the covariance and the uncertain information, and the uncertainty can be effectively processed by adopting the Riemannian center.