47 results about "Fuzzy uncertainty" patented technology

The invention discloses an inversing global SMFC (sliding mode fuzzy control) method for a micro-gyroscope based on a neural network. An inversing global SMFC system based on the neural network consists of an inversing global sliding mode fuzzy controller, a neural network dynamic characteristic estimator, and a fuzzy uncertainty estimator. Global sliding-mode control can iron out a defect that an arrival mode in the conventional sliding-mode control does not have robustness, speeds up system response, and enables a system to have robustness in the whole process of response. The method enables a lyapunov function of the system and the design process of a controller to be systematic and structuralized through inverse design during the inversing control. Fuzzy control is used for approximating a switching function term, and a switching term in sliding-mode control is converted into continuous fuzzy control output, thereby weakening vibration in sliding-mode control, and achieving a stronger capability of adaptive tracking. Therefore, the method improves the transient performance and robustness of a sliding-mode control system, estimates the unknown dynamic characteristics of the micro-gyroscope, and reduces the vibration in the control of a sliding-mode control structure.

The invention relates to a reliability optimization method of a distributed power supply system. In the method, the random uncertainty of a load forecasting value and the fuzzy uncertainty of the renewable energy type DG (Distributed Generation) output are mainly considered, the load forecasting value is represented by a random variable in normal distribution, the renewable energy type DG active output is represented by a trapezoidal fuzzy variable, and the minimization of an expected value of the sum of the investment and operating cost and the outage loss compensation cost of the distributed power supply system is taken as a target function, so as to establish a hybrid chance-constrained programming model based on reliability optimization. In comparison with the scheme obtained by the conventional method, the method provided by the invention has stronger adaptability to the uncertainty of future factors, and can largely reduce compensation investment after the event and reduce loss and waste.

The invention discloses a reliability design method of a force applying component of a high-speed press considering multiple types of uncertainties. The method includes the following steps: considering the random, interval and fuzzy uncertainties of high-speed press force components, choosing the minimum reliability value under the influence of three uncertainties as reliability index, establishing the stochastic model of high-speed press force components, and establishing the stochastic model of high-speed press force components; interval-Fuzzy mixed reliability design model; according to theconservation principle of entropy and '3 sigma criterion', a simplified stochastic-interval reliability design model; adopting Latin hypercube sampling and cooperative simulation technology, the Kriging model of function and objective function is constructed. The simplified reliability design model is decoupled from the reliability analysis to form a two-loop optimization solution. The inner loopuses adaptive step-size iterative method for reliability analysis, and the minimum value of reliability index is obtained. The outer loop uses genetic algorithm to optimize the design vector, and judges the feasibility of the design vector according to the reliability analysis results. When the maximum evolutionary algebra or convergence threshold is reached, the optimal solution is output.

The invention relates to an interval value fuzzy set method for aging state evaluation of insulation paper of a transformer, and belongs to the fields of a power system and automation thereof. To solve the problem that big data related to insulation state evaluation of the distribution transformer is diversified, large in amount and complex in mutual relation, knowledge mining and inference principles are used on the basis of establishment of a big data library to process and analyze the big data; and aimed at random and fuzzy nondeterminant parameters related to insulation state evaluation ofthe distribution transformer, an interval value fuzzy set theory is used for processing and analysis, and the aging state of the insulation paper of the transformer is evaluated accurately.

The invention discloses a method and a system for calculating output power of a water cooling photovoltaic-solar thermal integrated power generation system based on three-dimensional trapezoidal fuzzy. The method comprises the steps of calculating a temperature reduction value of a photovoltaic power generation panel and calculating an output power increment value of the photovoltaic power generation panel; calculating a three-dimensional generalized trapezoidal fuzzy set of a multi-value fuzzy uncertainty relation of output power and sunlight intensity of a water cooling photovoltaic system, and calculating effective sunlight intensity of a predetermined region by utilizing the three-dimensional generalized trapezoidal fuzzy set; calculating generation power of the photovoltaic power generation panel of the water cooling photovoltaic system and an available high-temperature high-pressure steam quantity of the water cooling photovoltaic system; calculating an electrical efficiency increment value of a water cooling solar thermal system; calculating generation power of the water cooling solar thermal system; and calculating the output power of the water cooling photovoltaic-solar thermal integrated power generation system. The output power of the water cooling photovoltaic-solar thermal integrated power generation system can be accurately predicted, so that necessary technical support is provided for distributed new energy power generation and intelligent power grid dispatching operation.

The invention relates to an information diffusion-based small watershed mountain torrent disaster risk analysis method. According to the method, an information diffusion theory is introduced; based onlimited historical mountain torrent data, set-valued processing is performed on samples with monodromy observation values under incomplete information conditions, so that the samples can be transformed into fuzzy set-valued samples with fuzzy uncertainties; a mountain torrent disaster information matrix is established; the calculation of Pearson's III type curve, exceedance probability and confidence is introduced; and therefore, the quantitative assessment of a mountain torrent disaster risk can be carried out rapidly, and a result is closer to actual situations. With the method of the invention adopted, the objectivity and scientificity of the analysis result of the mountain torrent disaster risk can be improved.

The invention relates to an attack detection method based on fuzzy uncertainty reasoning, which belongs to the technical field of network information security. Suitable for misuse detection systems. On the basis of extracting fuzzy attack features and establishing fuzzy knowledge templates, the present invention uses fuzzy reasoning technology to detect the possibility of attack occurrence. The misuse detection system adopts the method of the invention, which can effectively improve the detection accuracy of attack behavior and variants thereof.

The invention discloses a power system comprehensive decision-making method based on double-layer fuzzy optimization, belonging to the field of power system management. The invention provides a hierarchical decision-making method for electric power production and environmental management in electric power system planning, considers the fuzzy uncertainty of economic parameters in the system and uses a double-layer fuzzy programming method to solve the problem. In the process of solving, the decision makers at different levels control different decision variables and optimize their own objectives. Decision-makers at the upper level make decisions first, and the lower level reacts according to the decision-making results at the upper level. Therefore, the optimization method disclosed by theinvention obtains the optimal power system planning scheme by balancing the benefits of the two layers, and the optimization result disclosed by the invention can provide technical support for studying the regional power system planning, the comprehensive utilization of resources and the environmental protection.

The invention discloses a dynamic demand response pricing method based on fuzzy reinforcement learning. The method comprises the following steps: S1, establishing a hierarchical power market model, comprising a fuzzy load demand response model, a load aggregation quotient optimization model and an objective function model; S2, the model established in step S1 is solved by the fuzzy reinforcement learning algorithm to obtain the optimal retail price. The invention searches for the reasonable electricity price under the condition of considering the fuzzy uncertainty of the load response, Aimingat the shortcoming that the fuzzy uncertainty of load response is not taken into account in the dynamic demand response pricing model, a fuzzy load demand response model is proposed, load aggregationquotient optimization model and objective function model, A dynamic demand response pricing method based on fuzzy reinforcement learning is proposed, which not only fully considers the uncertainty ofload response, but also adapts to the dynamic power market environment and improves the computational efficiency. By optimizing the real-time optimal pricing strategy, the reliability of power systemcan be improved and the energy imbalance can be reduced.

The invention discloses a distribution network planning method based on a credibility theory, which comprises the steps of S1, building a distribution network fuzzy planning model based on the credibility theory; S2, building an equivalent model with a given reliability index; and S3, solving the model by adopting fuzzy power flow calculation and a genetic algorithm. According to the invention, the distribution network planning model based on the credibility theory is built on the basis of a fuzzy mathematic theory, the fuzzy uncertainty of load is considered, the minimization of the fixed investment and the fuzzy expected loss cost of the distribution network within the planning period is taken as an objective function, the reliability index is introduced, and the branch power and node voltage constraints are processed by adopting fuzzy change constraints; and secondly, the distribution network planning model under the credibility index is equivalent to a distribution network planningmodel under interval load by adopting a credibility index equivalence theorem, and a cut set corresponding to the fuzzy load distribution is enabled to be expressed by an intuitive interval. In addition, the solving method lowers the model solving difficulty and reduces the model solving time.

The invention relates to the technical field of reliability analysis, and provides a knuckle bearing reliability analysis method considering fuzzy uncertainty. The method comprises the following stepsof: establishing a performance function of a joint bearing for expressing uncertainty by using a random variable and a fuzzy variable; solving a membership interval of the fuzzy variable under a preset membership level; and calculating a reliability membership function and a maximum stress point stress mean membership function of the joint bearing according to the random variable, the preset membership level, the membership interval and the performance function. According to the method, a random variable and a fuzzy variable are considered while a performance function is established; comparedwith the prior art, the reliability analysis is carried out from multiple aspects, the reliability of the knuckle bearing can be analyzed more comprehensively, and guidance can be provided for subsequent optimization design of the knuckle bearing.

The invention discloses a structural fuzzy uncertainty analysis method based on self-adapting matching points and belongs to the field of structural uncertainty analysis. Firstly, an interval of a membership degree axis is obtained; secondly, a cut set interval of fuzzy uncertain variables is obtained through a fuzzy cut set policy; thirdly, by means of an interval uncertainty analysis method, upper and lower structural response boundaries of the cut set interval are obtained; finally, by means of a two-point trapezoid formula and a three-point trapezoid formula, the area defined by the upperand lower structural response boundaries and the membership degree axis is calculated, whether or not the area is converged is judged, structural response fuzzy distribution is output if yes, or if not, next circulation is carried out until the area is converged. It is shown by a numerical example that by means of the structural fuzzy uncertainty analysis method based on the self-adapting matchingpoints, precise structural response fuzzy distribution can be obtained, and a new approach is provided for structural fuzzy uncertainty analysis.

The invention relates to a fault fracture zone tunnel gushing water disaster occurrence probability grade evaluation method. The method comprises the steps of designing index system constituent elements; establishing a water inrush disaster occurrence probability evaluation index system; obtaining each level membership degree of the evaluation index of the occurrence probability of the water inrush disaster by utilizing an expert evaluation mechanism; carrying out weight calculation on the evaluation indexes of the occurrence probability of the water inrush disaster; and evaluating by adopting a multi-level fuzzy comprehensive evaluation method. The method overcomes the defect of subjectivity of a decision result of an analytic hierarchy process in existing disaster risk evaluation, adopts a fuzzy comprehensive evaluation method to make a decision, describes and quantifies fuzzy uncertainty factors by utilizing concepts of a fuzzy set and membership, weakens subjectivity of evaluation by adopting an entropy weight method, and the defect of relatively high uncertainty in evaluation of the occurrence probability grade of the water gushing disaster in the tunnel is overcome.

The invention belongs to the technical field of modeling of daily net active powers of high wind power penetration busbars, and relates to a novel modeling method of considering that the net active power of each busbar has dual uncertainty characteristics of randomness and fuzziness. The method comprises the following steps: selecting daily net active power data of the busbars in a specific region over the years, fitting the probability distribution of the net active power data within each time period divided according to the requirements when the confidence ratio is 0.05 by a MATLAB probability density curve, and determining a confidence interval and a membership function model of the parameter fluctuation range; defining the net active powers as random fuzzy variables, obtaining an opportunity measurement function of the net active powers and building a random fuzzy uncertainty model; and finally providing the process and step of simulation generation of simulated daily net active powers of the busbars through a random fuzzy simulation technology and an inverse transform method. The novel modeling method can relatively comprehensively describe multiple uncertainty characteristics of the daily net active powers of the high wind power penetration busbars, and provides a corresponding guiding basis for electricity generation scheduling and dispatching of a high wind power-integrated power system in the future.

The embodiments of the invention disclose a daily generation fuzzy probability calculation method and device for a photovoltaic power generation system, which are used for solving the technical problems that the daily generation calculation method for a distributed photovoltaic power generation system does not comprehensively consider uncertainty and randomness of influence factors and the applicability, practicability and application of the calculation method are difficult to meet in the prior art. The method comprises: according to daily generation data of the photovoltaic power generation system, calculating an n-dimensional generalized trapezoidal fuzzy set for determining daily generation fuzzy uncertainty of the photovoltaic power generation system; calculating an n-dimensional generalized trapezoidal fuzzy set of time period generation of the photovoltaic power generation system according to the influence of sunshine intensity, sunshine temperature rise, sunshine shadow and sunshine deflection angle on the photovoltaic power generation system; calculating the daily generation of the photovoltaic power generation system according to the n-dimensional generalized trapezoidal fuzzy set of time period generation of the photovoltaic power generation system; and determining the daily generation fuzzy probability of the photovoltaic power generation system by calculation according to the daily generation of the photovoltaic power generation system.

The invention discloses a distribution network random fuzzy power flow algorithm based on two-stage random fuzzy simulation, and belongs to the field of uncertain power flow calculation of power systems. The method considers the impact of two kinds of uncertainties: random and fuzzy of wind power output, photovoltaic output and load demand on network state variables. Firstly, a random fuzzy uncertainty model of wind power output, photovoltaic output and load demand is established, and then a distribution characteristic model of system voltage and power is obtained by two-stage random fuzzy simulation technology. The multi-point estimation technique is utilized in the random simulation phase, and the fuzzy simulation technique is utilized in the fuzzy phase to improve the computational efficiency of the algorithm. The algorithm can simultaneously consider the effects of both random and fuzzy uncertainties in the system, and can adapt to more system environments and be compatible with algorithms such as system stability control.

The present invention relates to the field of power systems and automation of the power systems, and in particular to a transformer failure rate calculation method considering the temperature. The method comprises: based on the establishment of a large database, using the data clustering principle to process and analyze big data; and processing and analyzing parameters of random and fuzzy uncertainty involved in transformer failure state evaluation by using the theory of a probability fuzzy set. According to the transformer failure rate calculation method considering the temperature proposed by the present invention, the transformer failure state can be evaluated, the uncertainty of the transformer failure state characteristic value is reflected, the theoretical guidance is provided for the transformer failure state evaluation, and the necessary technical support is provided for the maintenance and operation of the distribution network.

The invention discloses a spectral pattern recognition method for finely recognizing a spatial target. The spectral pattern recognition method comprises the following steps: acquiring spectral data ofvarious spatial targets by adopting a spectrograph; Preprocessing the acquired spectral data to remove noise interference; Performing data feature extraction on the preprocessed data; performing Moderecognition on data obtained after feature extraction, wherin a modeling method is information entropy weight fuzzy rough neighbor (Entropy Weight Fuzzy-rough nearest Neighbour, referred to as EFRNN)method .According to the method, concepts of an information entropy weight and a fuzzy rough set are introduced; The information entropy weight considers all information of the sample and quantizes the information, and introduction of the fuzzy rough set can avoid fuzzy uncertainty caused by overlapping and uncertainty caused by insufficient characteristics to a certain extent, so that the pattern recognition precision of similar samples is improved. The method has the advantage that parameters do not need to be preset, and accurate distinguishing can be carried out under the condition that samples are similar.

The invention discloses a reliability analysis method based on hybrid uncertainty variables and application thereof, and the method comprises the following steps: performing variable processing, including interval variable processing and fuzzy variable processing, for interval variable processing, converting an interval variable x into a random variable U obeying standard normal distribution, fuzzy variable processing comprising fuzzy variable initialization and fuzzy variable set cutting processing; inputting a starting point; calculating uk+ 1; judging a KKT condition; judging convergence; calculating a maximum failure probability and a minimum failure probability under the current cut set; and judging a termination condition. According to the reliability analysis method based on the hybrid uncertainty variables, multiple uncertainty types can be optimized, hybrid reliability analysis with random, interval and fuzzy uncertainty variables at the same time can be completed, a brand-new technical processing thought is provided for hybrid reliability analysis, and the processing flow is greatly simplified. The technical level of reliability analysis of various types of uncertainty is improved, and the method has great application prospects.