43results about How to "Take full account of uncertainty" patented technology

The invention relates to the field of planning of power transmission networks of electric power systems. For assisting planning personnel of the power transmission network to evaluate a power grid capacity margin in the future, discriminate system weak link recognition and an unreasonable grid structure, balance reliability and economy of a power grid planning scheme, and the like, the invention provides a calculating method of a utilization rate of an electric transmission line based on probability load flow. The technical scheme includes that the calculating method comprises the following steps: 1), carrying out definition description on the utilization rate of the electric transmission line; 2), building a probability distribution model of loads of N years in the future and a generator shut-down probability model; 3), generating random loads and generator random shut-down states; 4), serving lowest cost of total output of each generator set as an economic distribution objective function, and satisfying constraint conditions at the same time; 5), calculating line load flow values through a direct current load flow method or an alternate current load flow method; and 6), recoding the line load flow values, obtained by simulation everytime, of the line. The calculating method is mainly applied to the planning of the power transmission networks of the electric power systems.

The invention relates to a power system distribution transformer load data abnormity detection and repair method, and belongs to the technical field of power system analysis and control. The method comprises a coarse detection stage and a micro detection stage of the abnormal condition of the distribution transformer load data: in the coarse detection stage, missing data is complemented by using alow-rank matrix technology, and an obvious abnormal value is preliminarily repaired; In the micro-detection stage, a random forest quantile regression model is used for constructing the influence relation of factors such as historical loads, weather and day types on the to-be-detected loads, and local abnormal data are finely adjusted. According to the method, diversified data and historical loaddata can be fully utilized to detect and repair abnormal data in the distribution transformer load in the power system, a reliable data basis is provided for power load prediction and power system operation, the power load prediction precision is effectively improved, and the operation cost of the power system is reduced.

The invention provides a building outer surface wind pressure measurement method. The method includes the steps that a rigid model is manufactured and pressure measurement points are set; a wind tunnel test is conducted on the rigid model in a single round, so that a point wind pressure coefficient independent peak value reserving sequence and a face average wind pressure coefficient independent peak value reserving sequence are acquired; maximum likelihood estimation is conducted on point scale parameters and point position parameters of the point wind pressure coefficient independent peak value reserving sequence and face scale parameters and face position parameters of the face average wind pressure coefficient independent peak value reserving sequence, so that the maximum design value and the minimum design value of point wind pressure coefficients and the maximum design value and the minimum design value of face average wind pressure coefficients are determined, wherein the point scale parameters, the point position parameters, the face scale parameters and the face position parameters submit to extreme value distribution; the most disadvantageous maximum design value and the most disadvantageous minimum design value of the point wind pressure coefficients and the most disadvantageous maximum design value and the most disadvantageous minimum design value of the face average wind pressure coefficients are determined; the wind pressure of the outer surface of a building is determined. The method has the advantages of being low in cost, high in safety and capable of being widely applied to the field of buildings.

The invention discloses a recommendation system noise filtering method based on information entropies and fuzzy C-means clustering. The method comprises steps that first, user historical scoring dataof a target recommendation system is collected and arranged; second, Monte Carlo stochastic simulation is utilized to construct sub data sets of the user scoring data, a recommendation algorithm is utilized to acquire information entropies and recommendation precision of different sub data sets; third, the information entropies are classified according to uncertainty levels, recommendation precision is classified according to recommendation precision levels, and an empirical model is constructed to determine the potential natural noise data proportion; fourth, fuzzy clustering analysis on allthe user scoring data sets is carried out, and noise data is identified and deleted; and fifth, a recommendation algorithm operates for all the scoring data sets, and a recommendation precision indexis utilized to evaluate recommendation quality. The method is advantaged in that quantization measurement of the user scoring information can be realized, and the proposed natural noise data filteringtechnology has certain universality and portability.

The invention belongs to the field of optimal dispatching in electric systems and discloses a coordinating optimal dispatching method for large water electricity, small water electricity and wind electricity. The method is characterized in that absorbing the largest electric quantity serves as the purpose, the confidence degree of an objective function and the partition output out-of-limit chance constraint are comprehensively taken into account, and then a coordinating optimal dispatching model for largest short-term output electric quantity of the large water electricity, the small water electricity and the wind electricity based on a chance constraint planning is constructed; power generation predict error distributions of the wind electricity and the small water electricity are obtained by the adoption of a nonparametric kernel density estimation method; output fluctuations of the wind electricity and the small water electricity are balanced by utilizing good adjustment capacity of the large and middle water electricity, and solving is performed by the adoption of an improved particle swarm optimization method and combined with Monte Carlo simulation. The method can fully take uncertainty of the small water electricity and uncertainty of the wind electricity into account, effectively coordinate large water electricity energy, small water electricity energy and wind electricity energy, adjust the remaining to supplement the lack, preferentially absorb the small water electricity and the wind electricity, provide a balancing basis of operation risks and expected absorbed electricity quantity for a dispatcher, and provide a reference for input of other types of large-scale intermittent energy.

The invention discloses a contract electric quantity decomposition method based on distribution robust optimization, which comprises the following steps: S1, collecting cost information, difference information between monthly power generation quantity and contract electric quantity, unit parameter and power data, and establishing a contract electric quantity decomposition model according to the difference information; S2, a distributed robust optimization model considering wind power uncertainty is established according to the contract power decomposition model; S3, according to the distributed robust optimization model, adopting the column and constraint generation algorithm to obtain the probability value result of the wind power uncertain scene; 4, accord to that probability value result, adopting the distribute robust optimization model to obtain the decomposition result of the contract electric quantity. The invention solves the problem of wind power and the uncertainty of load inthe decomposition of contract power quantity, and reduces the fluctuation caused by the uncertainty of wind power.

The invention relates to an evaluation method considering influence of new energy grid connection on the utilization rate of transmission network. The evaluation method comprises that a photovoltaic output probability model considering influence of illumination randomness is established; a wind power output probability model considering influence of wind speed randomness is established; a load probability distribution model and a generator outage probability model are established; the trend probability distribution of lines and transformer with and without new energy is calculated respectively in a Monte Carlo probability trend method; a transmission network utilization rate index system considering the operation characteristic is defined; and the utilization rate of the transmission network is calculated via the transmission network utilization rate index system, the utilization rate indexes of the transmission network with and without new energy are compared, and influence of new energy grid connection on the utilization rate of the transmission network is obtained. The evaluation method can be used to reflect the real capacity utilization condition of lines in future, search weak links and unreasonable rack structures in the electrical network system, evaluate the utilization state of the future electrical network more accurately, and consider the influence of new energy grid connection in a concise and effective manner.

The present invention provides a CDM (Coefficient Diagram Method)-based missile autopilot design method. The design method comprises: figuring out an open-loop characteristic polynomial P0(s), and drawing a coefficient map of the open-loop characteristic polynomial P0(s) based a coefficient map drawing method in the CDM; determining the structure of an autopilot according to the convexity of curves in the coefficient map of the open-loop characteristic polynomial P0(s), and obtaining a closed-loop characteristic polynomial P(s) having autopilot control parameters; selecting a stability index [gamma]i and an equivalent time constant tau, and figuring out a closed-loop characteristic polynomial P(s) of the system; and equating coefficients of corresponding orders of the two closed-loop characteristic polynomial P(s), and working out the autopilot control parameters. The method provided by the invention is applicable in the design of the missile autopilot, and can realize online setting of the parameters of the autopilot, providing simple operation and convenience in implementation, and achieving satisfactory effects in overload instruction tracking and anti-interference performance.

The embodiment of the invention discloses a reservoir dispatching risk evaluation method by considering runoff forecast uncertainty. The reservoir dispatching risk evaluation method comprises a data processing module, a runoff prediction module, a runoff probability distribution module, a runoff random sampling module and a risk evaluation calculation module, wherein the data processing module standardizes data required by the runoff prediction module and inputs the data into the runoff prediction module to carry out runoff forecast; the forecasted runoff data is input the runoff probability distribution module by the runoff prediction module to obtain a corresponding probability distribution matrix; the probability distribution matrix is input into the runoff random sampling module by the runoff probability distribution module to carry out random sampling to obtain a storage runoff set; and the storage runoff set is input the risk evaluation calculation module by the random sampling module to calculate module calculation risks. The invention fully considers runoff forecast uncertainty and randomness, establishes a reservoir dispatching risk evaluation model by considering the runoff forecast uncertainty, carries out comprehensive safety assessment on the dispatching operation of a reservoir under runoff forecast information and exhibits wide application prospect and social value.

The invention discloses a version test method and device. The version test method comprises the following steps: respectively assigning corresponding parallel versions for multiple test groups with the same number as test users; determining a to-be-selected version according to the test situation of each of the parallel versions; increasing a new test user for the test group corresponding to the to-be-selected version so as to re-determine a new to-be-selected version according to the test situation of each of the parallel versions, and increasing a new test user for the test group corresponding to the new to-be-selected version till cyclic termination conditions are met; determining the parallel version corresponding to each of the test groups with the largest number of test users into an optimal version. Compared with an existing method for determining the optimal version through analyzing disposable data acquisition and use, the version test method disclosed by the invention has the benefit that the uncertainty of user preferences is fully considered, so that the determined optimal version is usually more accurate, thereby solving the problems in the prior art.

The invention discloses a device fault diagnosis method and a device fault diagnosis system based on a Bayesian network. Wherein the method may be based on a known fault of an input device to be diagnosed, a physical model of a corresponding device and failure condition probability information of the training data set after association rule mining are retrieved from a memory, the optimal fault diagnosis recommendation under the known fault information is given and the optimal fault diagnosis is stored and the optimal fault diagnosis is stored, fully considering the uncertainty of information,the diagnosis results are the optimal solution under the current conditions, not the qualitative value. The database can be updated according to the actual diagnosis results to ensure the accuracy ofdiagnosis.

The invention discloses a method and a device for determining the proton treatment dosage and range by using proton induced thermoacoustic signals. The method includes: receiving a proton induced thermoacoustic signal through an ultrasonic sensor, and analyzing the proton induced thermoacoustic signal and extracting the travel time and waveform information; performing ray tracing on a propagationpath of the proton induced thermoacoustic signal in a body, and calculating the minimum travel time of a first arrival wave; using the perturbation method to determine a partial derivative matrix of the travel time to a position of a thermal sound source, and establishing an inversion equation; obtaining the position of the thermal sound source and a velocity model through iterative fitting; solving the problem of double coupling of thermal sound source position disturbance and velocity model disturbance by using the parameter separation method; further correcting the position of thermal soundsource and the velocity model that the proton induced thermoacoustic signal is propagated in human body through the disturbance; and determining the dosage and the range of proton treatment through aproton beam energy distribution relationship. The experimental results obtained by the invention not only save the running time but also ensure the positioning accuracy; and the method and the devicecan be widely applied in proton treatment.

The invention relates to a tunnel surrounding rock geological information prediction method based on built tunnel information intelligent identification, and the method comprises the steps: carrying out the classification and storage of engineering geological information through data preprocessing and formatting, and carrying out the data screening of built tunnel information through a clustering algorithm; establishing a BP neural network prediction model on the basis of data screening, carrying out training, and verifying the reliability of the model; and finally predicting surrounding rock geological information of an unknown section of the tunnel by using the completely trained model. The tunnel surrounding rock geological information prediction model based on built tunnel information intelligent identification has good universality and high prediction precision, can fully consider built tunnel engineering surrounding rock data, and carries out deduction and dynamic update on tunnel surrounding rock geological information, therefore, more reliable surrounding rock geological information is obtained for fine analysis and design of tunnel engineering, and the method is suitable for prediction of most tunnel surrounding rock geological information.

The invention discloses a tie line power safety domain characterization method considering new energy uncertainty. The tie line power safety domain characterization method mainly comprises a tie linepower safety domain characterization method considering new energy uncertainty based on a probability density function and the tie line power safety domain characterization method considering new energy uncertainty based on an interval number. According to the method, the uncertainty of new energy can be fully considered, and the characterization precision of the joint line power safety domain isimproved.

The invention relates to a wind / light / storage / firewood microgrid optimal configuration method considering a meteorological fluctuation characteristic difference typical scene. The method is characterized by comprising the following steps: 1) analyzing wind-light-load time sequence fluctuation difference characteristics; 2) wind-light-load fluctuation characteristic difference and time sequence correlation considering wind-solar load fluctuation characteristic difference and time sequence correlation Generating a charge typical scene set; 3) wind / light / storage / firewood microgrid optimal configuration modeling considering a meteorological fluctuation characteristic difference typical scene, and 4) adopting a teaching and learning optimization algorithm TLBO to solve an optimal configurationscheme for the model in the 3), and has the advantages of scientific and reasonable method, high applicability, good effect, full consideration of speed uncertainty, capability of improving microgridplanning efficiency and the like.

The invention discloses a waterlogging storage water surface rate partition processing method based on urban waterlogging basic information. The method comprises the steps that basic information suchas rainfall runoff coefficients, main duration rainfall statistical results and current water surface rates, input by a user, of a current region is acquired; obtaining a designed flood drainage flowaccording to the basic information and the rainfall flood process control information; obtaining a designed regulation and storage water depth according to the designed drainage flow and the basic information; obtaining a corresponding designed rainstorm duration according to the designed flood drainage flow or the designed regulation and storage water depth; designing rainstorm duration accordingto the designed flood drainage flow to obtain partition indexes; zoning threshold delimiting is conducted on the waterlogging storage water surface rate of the current region through zoning indexes,and a region division result is obtained; and outputing results of zone partition. According to the processing method, the calculation process can be simplified, the workload of multi-scenario adjustment is reduced, and the operability is high; and the design scheme of the urban flood drainage system can be optimized according to local conditions in combination with the partitioning result of theurban water surface rate.

The invention discloses an evidence retrieval and collection system and method applied to evidence-based medicine. The evidence retrieval and collection system comprises an evidence collection unit, an evidence chain generation module, a data analysis module, an intelligent learning module, an evidence retrieval unit and a model construction module. The invention further discloses the evidence retrieval and collection method applied to evidence-based medicine. The evidence retrieval and collection method comprises the following steps: A, evidence collection; B, evidence chain generation; C, evidence retrieval; D, data analysis; and E, model construction. According to the invention, case evidences of patients can be fully collected and are combined with historical case evidences retrieved from a Cochrane database; timestamps and geographical location information are added into the collected evidence data through an information adding module to form an evidence chain; corresponding historical case evidences can be quickly retrieved through the Meta analysis module; and keyword retrieval is learned through the intelligent learning module, and a retrieval strategy of the intelligent learning module is learned and analyzed, so the purpose of intelligently making a retrieval strategy after related keywords are input is achieved.

The invention discloses a moving target positioning solution and error evaluation method under a noisy network, which comprises the following steps of: installing a label on a moving target, determining a sensing base station set corresponding to the label, and solving a differential distance based on a signal arrival time difference between a base station and the label in combination with a target motion model; a primary distance estimation value between a moving target and a base station under a noise signal is estimated by adopting unscented particle filtering, a target distributed positioning calculation model is established based on a base station calibration error and signal ranging, and a covariance matrix and a corresponding positioning error caused by uncertainty in the base station calibration and signal measurement process are calculated. According to the method, uncertainty between a base station and a signal is fully considered, unscented particle filtering is introduced to carry out primary positioning estimation on the moving target, in order to avoid positioning errors caused by an ill-conditioned matrix, a primary distance estimation value and a variable separation method are utilized to solve the real-time position of the moving target, positioning sensing is carried out on the moving target in real time, and the positioning errors are evaluated; and a basis can be provided for cooperative positioning and autonomous movement of the target.

The invention relates to the technical field of independent micro-grids, in particular to an independent micro-grid design system and method based on random robust hybrid optimization, and the systemcomprises a data collection module, an investment planning calculation module and a design scheme output module. The data acquisition module is used for acquiring design factor information; the investment planning calculation module receives the design factor information from the data acquisition module and analyzes the design factor information to generate a design scheme result; and the design scheme output module is used for exporting a design scheme result generated by the investment planning calculation module. According to the independent microgrid two-stage random robust investment planning model provided by the invention, uncertain parameters can be adjusted, so that the uncertainty of renewable energy sources and loads in the system can be fully considered, and then planning schemes under different adjustment parameters are obtained; therefore, an independent microgrid design scheme with more reasonable investment planning can be obtained.

The invention discloses an entropy method for determining concrete-filled steel tube bearing force design error distribution. First, concrete-filled steel tube bearing force design error distribution is depicted through candidate distribution, then the standard entropy and the forecast entropy of bearing force design errors are computed respectively, and finally through deviation indexes, the best distribution of the concrete-filled steel tube bearing force design errors is determined. According to the entropy method, the distribution of the concrete-filled steel tube bearing force design errors is selected, the uncertainty of bearing force error design data is fully considered, technology features are simple, and engineering technicians can use the method conveniently. The method can be used for forecasting the concrete-filled steel tube bearing force design errors, then the confidence interval of bearing force is estimated, the reference for designing a concrete-filled steel tube combination engineering structure is achieved, and great theoretical and engineering significance is achieved.

The invention provides a multi-time-scale smart energy station cooperative operation control method and device, and is used for carrying out the optimal scheduling of a smart energy station in consideration of renewable energy consumption, and the method comprises the steps: building an optimal scheduling model of the smart energy station through considering the operation cost and the green certificate-carbon transaction cost; dividing the operation of the smart energy station into three scales of day-ahead optimization scheduling, rolling optimization scheduling and real-time control, and respectively formulating specific control algorithms; according to the equipment characteristics of the smart energy station, the equipment in the station is divided into three types according to the start-stop flexibility, and participates in control of different time scales with different margins; and finally, monitoring the running state of the smart energy station in the running day, and solving the optimal scheduling model in a rolling manner to obtain an optimal scheduling plan of the smart energy station. The problems that renewable energy consumption and carbon emission reduction cannot be fully excited, and renewable energy output uncertainty and comprehensive energy system flexibility are not fully considered are solved.

The invention discloses a security patrol resource deployment method and system, and belongs to the technical field of security. The problem that limited security resources cannot be reasonably deployed by adopting an existing method is solved. The reward function of the defender is defined according to the reward and punishment conditions of the defender and the attacker, the reward function value is used for prioritizing the patrol targets, and the behavior of the attacker is predicted according to the quantum response model, so that the rational opponent hypothesis in the existing research is improved. In this way, the priority of the protection target and the uncertainty of the attacker are fully considered, the microscopic strategy is formulated based on the priority of the protection target and the uncertainty of the attacker, and a complete patrol deployment plan is expanded on the basis of the formulated microscopic strategy, so that limited security resources are fully scheduled and reasonably deployed, and the efficiency of the patrol system is effectively improved. The method and system can be applied to the technical field of security and protection.

A generation method of a power generation scheme for preventing sustained damage of a hydropower station group includes the steps of 1, determining an optimization target and constraint conditions and establishing a statin optimizing and scheduling module; 2, considering uncertainty of hydrological forecasting and conversion of forecasting runoff; 3, calculating potential output of each phase; 4, determining a pattern of power generation constraint rules; 5, acquiring optimized power generation constraint rules and a power generation scheme set by means of an optimization algorithm; and 6, establishing an evaluation index system to evaluate optimal scheduling results. The generation method has the advantages that sustainability of the power generation process is ensured, a minimum of damage is ensured, and the simple and universal power generation constraint scheduling rules are provided to guide optimization of hydropower stations; the influence of hydrological forecasting errors upon scheduling results is fully considered, the power generation constraint scheduling rules are optimized by means of forms of energy, and more accurate decision basis is provided for decision makers.

The invention relates to a method for predicting the spatiotemporal trajectory of a moving object in an obstacle space, which includes the following steps: simplifying the obstacle space by using the regional relationship between obstacles; presetting the expected distance between two objects in the obstacle space, and The obstacle space trajectory data is preprocessed to realize trajectory clustering and excavate the historical hotspot areas of moving objects; based on the Markov model, each cluster represents a hotspot area, and is judged according to the obstacle distance between hotspot areas and historical access habits The motion law of the moving object is obtained, and the trajectory prediction is completed by taking all the historical hot spots of the moving object as future candidate positions. The invention can accurately and efficiently analyze and predict the space-time track data of the moving object in the obstacle space, and the prediction result obtained by the invention is accurate and reliable.

The invention discloses a reactive power optimization method for a regional power grid based on limit power flow, which includes: collecting power information of the regional power grid for power flow calculation, and obtaining a power flow calculation result; obtaining a voltage deviation rate and a gateway power factor according to the power flow calculation result; determining The weight coefficient of the voltage deviation rate and the gate power factor, and obtain the weighted sum of the voltage deviation rate and the weight coefficient of the gate power factor; judge whether the weighted sum is greater than the engineering threshold. The present invention can adapt to the reactive power allocation requirements under different load modes, and adopts the PQ-PV node type conversion mode and the mode of relaxing the transformer stall constraint according to the characteristics of the feasible domain of the optimization problem under different load modes, and then expands the reactive power configuration, etc. The optimization space of the problem ensures that the reactive power configuration of the power system can fully meet the load demand in theory.

The invention relates to the field of planning of power transmission networks of electric power systems. For assisting planning personnel of the power transmission network to evaluate a power grid capacity margin in the future, discriminate system weak link recognition and an unreasonable grid structure, balance reliability and economy of a power grid planning scheme, and the like, the invention provides a calculating method of a utilization rate of an electric transmission line based on probability load flow. The technical scheme includes that the calculating method comprises the following steps: 1), carrying out definition description on the utilization rate of the electric transmission line; 2), building a probability distribution model of loads of N years in the future and a generator shut-down probability model; 3), generating random loads and generator random shut-down states; 4), serving lowest cost of total output of each generator set as an economic distribution objective function, and satisfying constraint conditions at the same time; 5), calculating line load flow values through a direct current load flow method or an alternate current load flow method; and 6), recoding the line load flow values, obtained by simulation everytime, of the line. The calculating method is mainly applied to the planning of the power transmission networks of the electric power systems.

The invention provides a damage prediction method and device for an energy system, equipment and a storage medium, and aims at mining time-space characteristics of typhoon wind speed through typhoon historical data and calculating wind speed factors under different time-space conditions to obtain a typhoon attenuation rule. And then, according to the wind speed factor, the predicted typhoon intensity and the vulnerability curve of the component, calculating the component failure rate of the integrated energy system under different time-space conditions. On the basis, the damage probability is proposed, and the damage probability under different time and space is calculated according to historical typhoon data so as to budget the damage probability under the current predicted typhoon intensity. According to the damage prediction method of the energy system provided by the invention, the uncertainty in the energy system is fully considered, the uncertainty is dealt with by integrating the flexibility of historical data in the energy system, and the obtained planning result can effectively deal with the uncertain fluctuation in the system, so that the prediction of the damage condition is more accurate and reliable.

The invention discloses a Bayesian network-based equipment fault diagnosis method and system. Wherein, the method can retrieve the physical model of the corresponding equipment from the memory and the failure condition probability information of the training data set after association rule mining according to the known faults of the equipment to be diagnosed, and give the known faults. The optimal fault diagnosis under the fault information recommends and stores the optimal fault diagnosis and stores the optimal fault diagnosis, which can fully consider the uncertainty of the information, and the diagnostic results given are the best under the current conditions. Solutions, rather than qualitative values, can update the database according to the actual diagnosis results to ensure the accuracy of diagnosis.