56results about How to "Guaranteed global optimality" patented technology

The invention discloses a driverless vehicle path planning method and device. The driverless vehicle path planning method comprises the following steps: S1, locating a current vehicle: obtaining the location of the current driverless vehicle as a starting point according to a differential GPS/INS system, then selecting a target point, loading the data of a global map, and reading the vehicle status information; S2, planning a global path: establishing a global raster map according to the coordinate information of the starting point and the target point, using a modified artificial potential field-ant colony algorithm to plan the global path, and generating a desired path; S3, driving and analyzing the current road condition by the vehicle, and analyzing and judging whether local path planning is needed or not according to the obtained sensor information; S4, selecting a local path planning strategy; and S5, sending the generated path information to a vehicle control layer. The invention improves the running efficiency of the program; and the vehicle can efficiently and reasonably plan a safe path in a complex and varied environment; therefore, the global optimality and local real-time performance of the path planning are ensured.

The invention relates to the technical field of wind power prediction, and discloses a wind power short-term prediction method. The method uses wind speed as an input, adopts a regression model of a least square support vector machine to predict output power of a wind power plant, and parameters of the regression model of the least square support vector machine are optimized by adoption of a chaotic particle swarm algorithm. The wind power short-term prediction method provided by the invention introduces chaotic motion characteristics into an iterative process, uses ergodicity of chaotic motion to improve a global searching capability of the algorithm in a searching process, overcomes the defects that the particle swarm algorithm is easy to fall into a local extreme point and is slow in convergence and low in precision in a later period of evolution, effectively solves the problem of prematurity of the particle swarm algorithm, can ensure global optimum, and achieves a better prediction effect; the method uses the least square support vector machine to predict, avoids the problem of solving quadratic programming, converts the prediction problem to a process of solving a linear equation set, and the solving process is greatly simplified; and the method adopts single wind speed as input data, and thus a prediction model is simpler.

The invention provides a hand-eye calibration parameter identification method, a hand-eye calibration parameter identification system based on a differential evolution algorithm and a medium. The hand-eye calibration parameter identification method comprises the following steps of moving a robot end of a robot vision system to different poses to collect robot joint data and camera image data; separately calculating a pose matrix of the robot end relative to a robot base coordinate system and a pose matrix of a calibration board relative to a camera coordinate system; defining a rotary component calibrated error function and a translational component calibrated error function; determining and solving a multi-target optimization function of a hand-eye calibration problem; and separately calculating calibrated errors of a rotating part and a translational part of the robot vision system and verifying and identifying the optimum hand-eye calibration parameter. The global optimality of a calibration result can be acquired, the acquired calibration result falls onto a special Euclidean group SE (3), and additionally introduced calculation of calibrating orthogonalization of the acquiredrotating matrix.

The invention discloses a path planning method for a foot-type robot. An improved A* algorithm and a self-adaptive DWA algorithm are utilized to plan a reasonable path of robot motion, one characteristic of the improved A * algorithm is as follows, as long as a passable path from a starting point to a target point exists in the environment, the method can always be successfully searched. Comparedwith other local path algorithms, the method has very outstanding advantages, the thought of the fusion algorithm adopted by the method is to perform global path planning by using the characteristic,and then perform real-time local planning in combination with the adaptive DWA, so a final path is close to global optimum and has better dynamic adaptive capacity at the same time.

The invention discloses a rotary machine residual life prediction method based on an integrated GMDH framework, and the method comprises the following steps: S1, collecting the data of a plurality ofsensors in a process from normal operation to fault failure of a plurality of rotary machines of the same type, and obtaining a training data set W through data processing; s2, dividing the data set differently to be used for constructing three GMDH prediction networks with differences respectively; s3, taking prediction output of the three GMDH networks on the training sample as input of a three-layer BP neural network to train the BP neural network, wherein the BP neural network is used for integrating prediction results of the three GMDH networks; and S4, predicting the residual life of therotary machine by utilizing the integrated GMDH framework, and calculating and outputting a residual life prediction value. Compared with a classic LSTM network and a single GMDH network, the methodcan effectively improve the prediction precision and generalization ability, and has greater practical guiding significance.

The invention discloses a quality soft sensing and monitoring method based on optimal nearest neighbor component analysis, aiming at solving how to optimize input characteristic variables related to each quality index from the angle of ensuring global optimization, and establishing corresponding quality index soft sensing and monitoring model based on the input characteristic variables. The methodof the invention considers the problem that each quality index corresponds to different input characteristic variables, and determines the optimal input characteristic variables corresponding to eachquality index according to the objective function value of the nearest neighbor component analysis algorithm by exhausting all possible combination forms of the input variables. The method of the invention utilizes the optimized input variable to establish the soft sensing model for each quality index, and utilizes the estimated value of the soft sensing model to monitor the status of the productquality in real time. The method of the invention can be said to be able to solve the soft sensing and monitoring problems related to the product quality.

The invention discloses a method for planning a field operation path of an unmanned tractor. The method comprises the following steps: A, preprocessing a farmland plot; B, acquiring position information of each vertex of the plot, and performing plot geometric modeling; C, planning a global path by adopting an offline global path planning algorithm, and generating an expected path; D, presetting asafe distance between the tractor and an obstacle before driving, driving the unmanned tractor according to the expected path and analyzing the current working environment in real time, and judging whether to plan a local path or not by combining a perception sensor to obtain obstacle information; E, calling an online local path algorithm for path planning, and allowing the unmanned tractor to travel according to the local path until the unmanned tractor avoids obstacles; and F, allowing the unmanned tractor to keep running on the expected path. The invention also discloses a device for realizing the method. According to the field operation path planning method and device for the unmanned tractor, the full-area coverage of a land parcel can be achieved, high universality and a good obstacle avoidance effect are achieved, and the re-ploughing rate and the missing ploughing rate can be effectively reduced.

The invention discloses a hybrid path planning method based on FLOYD and Astar. The method comprises the following steps: firstly building a vehicle mathematical model, constructing a high-precision map, combining a Floyd algorithm with an A * algorithm under the condition, and carrying out the path planning of an automatic driving vehicle through a hybrid algorithm. In a common path planning algorithm, the Floyd algorithm is high in time complexity, and a distance matrix and a path matrix obtained through calculation cannot meet the high real-time performance requirement of an automatic driving vehicle during dynamic driving; and the A * algorithm has a logic design problem, so that the planning result is not always globally optimal. Aiming at the problems of the Floyd algorithm and the A * algorithm in the traditional path planning algorithm, the method provided by the invention provides the hybrid path planning algorithm which combines the characteristics of the two methods, gives consideration to global optimality and real-time performance at the same time, enhances the adaptability of the algorithm to the environment, and improves the path planning efficiency. Therefore, online planning and dynamic global obstacle avoidance of an automatic driving reference path are realized.

The invention provides a time sequence progressive type reactive voltage optimal allocation control method. A progressive time sequence from day-ahead optimization to inter-day correction and then to real-time control is carried out, the dynamic characteristic of a system is considered during optimization process, monotonic segmentation of a load is carried out with a variation trend, the optimal target functions are different from different segments, and the economy and the safety of the system are jointly considered; and meanwhile, multi-period coupling solution is carried out by using a prime-dual interior point method based on non-linear complementarity. According to the method, rolling reactive power optimization control can be carried out according to the dynamic change trend of a power grid, the problem that the current power grid state is only and optimally controlled by automatic voltage control (AVC), so that the power grid runs at a more optimal state; meanwhile, reactive optimization control is more reasonable, and the economy, the safety and the reliability of power grid running are guaranteed; and the calculation convergence of a large-scale non-linear optimization problem in which a continuous variable and a discrete variable coexist and global optimum of solution are ensured, and an important guiding significance is provided for fining reactive voltage control of the power grid.

The invention relates to an electricity-gas interconnection system unit combination linear model and system considering electricity-to-gas coupling. According to the electricity-gas interconnection system, bidirectional coupling of the electricity-to-gas conversion technology and the gas turbine is considered, the unit combination problem of the gas interconnection system is studied, the lowest comprehensive operation cost of the whole system serves as the target, multiple safety constraints of an electric power system and a natural gas system are considered, and an electricity-gas interconnection system unit combination model is established and linearized to obtain a linear model. Through simulation calculation, the existing nonlinear model and the linear model are compared, the solving efficiency of the nonlinear model and the linear model is compared, meanwhile, the system operation cost and the operation state under the scenes of considering electricity-to-gas and not considering electricity-to-gas and electricity-to-gas are analyzed, and the result shows that the linear model improves the combination solving efficiency of the electricity-gas interconnection system unit, and the application of electricity-to-gas also helps to improve the economical efficiency of the electricity-gas hybrid connection system.

The invention discloses a trajectory control method for a vertical recovery landing section of a rocket. A new optimization problem is obtained by weighted summation of objective functions in two convex optimization problems, compared with a traditional guidance method, two optimization objectives are solved at the same time, the calculation amount is reduced while the optimization performance is ensured, a the difficulty in considering uncertainty and disturbance is also reduced; meanwhile, a convex optimization method and a robust model predictive control algorithm are combined, an original non-convex feasible region is converted into a convex feasible region, under the condition that atmospheric disturbance existing in the vertical recovery process is considered, global optimality of understanding is effectively guaranteed, and meanwhile, the defect of a model predictive control algorithm in the aspect of calculation time is overcome.

The invention provides a gradient separate zone optimization design method for a spacecraft pulse rendezvous trajectory, belongs to orbit design and optimization technologies in the field of spacecraft orbit dynamics and control, and in particular belongs to a rendezvous trajectory design technology in which a spacecraft rendezvous trajectory global optimization solution is obtained through calculation based on an interval arithmetic mathematical model. For a global issue of the method, a gradient optimization result is only used to determine an interval separation point and update a value of an upper bound of a target function, thereby preserving globalness of an interval optimization algorithm. For a large computation amount in the interval optimization algorithm, a gradient optimization method is introduced, so that the value of the upper bound of the target function can be quickly updated, and a gradient segmentation method can be used; in combination with a symbol segmentation method and an interval optimization strategy, intervals that do not comprise an optimal solution are effectively separated and removed, thereby improving operational efficiency. To solve the problem that the interval optimization algorithm has a high storage requirement, only a limited number of intervals are selected for processing in each iteration, so as to ensure that a storage space overflowing phenomenon does not occur in an operation.

The invention discloses a method for planning an individualized optimal riding tour route according to the requirement of a user. The method comprises the following steps: firstly taking the quality of a scenic spot into consideration, predicating the dynamic quality of the scenic spot according to an echo state network, acquiring the comprehensive quality of the scenic spot in combination with the static quality of the scenic spot; then taking the user experience quality in the riding process into consideration, planning an overview of a city, and characterizing the infrastructure of the city along the density of landmark buildings. Each directive scenic spot cluster is subjected to Q-value training and iteration until convergence through a Q-learning algorithm according to the user preference; and then the global optimal route is planned according to the iterated Q value. The invention further provides a specific knot inserting algorithm to meet the requirement of part of users hoping to have a few specific scenic spots in the route, the algorithm ensures that the global optimality of the route is not broken, and the tourism experience of riding of the user is guaranteed.

The invention discloses a power flow obtaining method applicable to an AC and DC system. The power flow obtaining method comprises the steps of: firstly, establishing an AC and DC power system power flow model including an AC network model, a DC network model and a converter model, then, relaxing the AC and DC power system power flow model to a second-order cone form by adopting a cone relaxationtechnology, and finally, on this basis, obtaining a piecewise linearization AC and DC power system power flow model according to a second-order cone linearization method, wherein the linearization precision can be adjusted through a zooming method and changing of a piecewise number index. According to the power flow obtaining method applicable to the AC and DC system disclosed by the invention, the cone relaxation technology and the second-order cone linearization technology are adopted; the original non-convex nonlinear AC and DC power flow equation is converted into the convex piecewise linearization form; the operation economy of the whole AC and DC system can be improved; and simultaneously, the power flow obtaining method is also more suitable for processing all kinds of operation optimization problems having relatively high timeliness requirements in the power system.

The invention relates to a power or static power-based reinforced concrete simply supported beam fire model correction method, and belongs to the technical field of model correction methods. Accordingto the method, damage feature parameters are selected as input parameters for model correction; physical parameters under a corresponding damage state are selected as output parameters for model correction; the damage feature parameters are combined by utilizing frequency and vibration mode changes; and when to-be-corrected physical parameters are selected, structural boundary conditions and material performance parameters, which influence structural mode information and damage recognition, are comprehensively considered. According to the method, a support vector machine statistical learningalgorithm is applied and a distributed correction strategy is adopted, so that sample counting in once correction is decreased, training samples are greatly decreased, the model correction correctnessis improved, and corrected finite models can reflect structural vibration characteristics and other structural responses more really in fire processes or after fires.

The invention provides a radon thorium daughter fixed energy region proportion coefficient deduction algorithm. The algorithm comprises the following steps of acquiring a spectral line; performing smoothing and spectral drift correction on the spectral line; determining a spectral line energy region; training a support vector regression machine; and performing radon and thorium daughter deductionon the actually measured spectral line by utilizing the trained support vector regression machine. A radon thorium daughter deduction effect is a main factor affecting a minimum detection limit of a manual radioactive aerosol continuous monitor; and the support vector regression machine has high generalization ability, so that the algorithm has the advantages of global optimality, theory for smallsample statistics, good fitting characteristics for nonlinear curves and the like. The radon thorium daughters are deducted by adopting the support vector regression machine; the performance of the algorithm is superior to that of average value, empirical formula and neural network deduction methods adopted at present; and the deduction accuracy of the radon thorium daughters is improved, so thatthe minimum detection limit of the monitor is further reduced.

The invention provides a black-start path generation and system recovery decision-making system based on a 3D visualization technology, which is used for power grid black-start and power grid rapid recovery after large-area power failure, and is characterized in that the decision-making system comprises a graphical interface module, a database and an algorithm function library; the graphical interface module comprises a graphical configuration software system and a 3D graphical system and is used for generating a plant station diagram, a grid frame plane diagram and a grid frame diagram of a 3D actual geographical background of a power system; the database can import power system data in BPA and CIME formats through a data conversion interface, and the power system data comprises element parameters and corresponding calculation results; the algorithm function library comprises a black-start scheme and an algorithm program for numerical analysis, simulation and optimization in a systemrecovery process so as to realize partitioning, black-start path generation, verification, evaluation and system recovery functions; according to the invention, the problem of heavy workload of calculation, analysis, writing and the like of a black-start scheme of a scheduling operator can be overcome.

The invention discloses a two-stage layering hybrid system of a separation module spacecraft. An operation control and mission planning portion comprises a decision management layer and a behavior executing layer. The decision management layer is provided with two management modules, sends cooperation requests for a specific mission, clears a cooperation mechanism and strategies, conducts the mission planning after the cooperative relationship is determined, and generates a specific mission command. The behavior executing layer is provided with a plurality of working modules, forms the cooperative relationship with the decision management layer in a certain mode, obtains the allocated mission, conducts the mission planning under the guidance of the cooperation mechanism and generates a motion sequence so as to enable the working modules to cooperatively complete the observing mission. One management module is set to be a standby module. The two-stage layering hybrid system of the separation module spacecraft has the advantages that the system suitability is good, the system calculated amount is reasonable and balanced, the CPU bottleneck problem does not easily occur, and the system stability is good.

The invention relates to a static optimization method of a distribution network based on photovoltaic storage charging tower access, which comprises the following steps: step 1, according to the historical data of photovoltaic system, predicting the output of photovoltaic system power generation; 2, presetting the capacity of the energy storage system and determining that the operation mode of theenergy storage system is peak clipping and valley filling; 3, establishing a charging load model of that electric vehicle according to the historical load data of the electric vehicle; 4, according to the model of the optical storage charging tower and the mathematical model of distribution network reconfiguration, establishing a distribution network reconfiguration model considering the access of the optical storage charging tower; Step 5, adopting the second-order cone programming method, converting the distribution network reconfiguration model considering the light storage charging toweraccess to obtain the distribution network second-order cone reconfiguration model under the light storage charging tower access, and solving the problem by GAMS. The invention can minimize the activepower loss in the daily operation of the distribution network and reduce the operation cost of the distribution network.

The invention discloses a reservoir operation chart optimization method based on dynamic planning. The method comprises the following steps that: dividing a whole operation period into T time periods; separating an operation line, and dispersing the operation line in a feasible water level range of each time period to obtain water level discrete points; establishing all water level discrete point combined spaces Ct<k> of the operation line, and determining a reservoir operation chart recursive computation model based on the dynamic planning; from the Tth time period, beginning to carrying out inverse time sequence recursive computation; enabling t to be equal to t-1, entering computation of a next time period, and when the computation of a (t-1)th time period is finished, ending an inverse time sequence recursive computation process; and from the first time period, carrying out recursion to the last time period in the time sequence, and determining an optimal discrete water level combination (Ct) to obtain an optimal reservoir operation chart. By use of the method, a reservoir operation chart drawing model is coupled with a dynamic planning model, an inverse time sequence recursive way is adopted to calculate the optimal reservoir operation chart, and the global convergence of the dynamic planning is fully utilized for guaranteeing the global optimality of the obtained operation chart.

A cold and hot gas combined supply dispatching method for a flexible area comprehensive energy system comprises the steps of: stepwisely introducing electricity-to-gas two-stage operation and the combined refrigeration technology of an electric refrigerator and a lithium bromide absorption refrigerator and carrying out modeling respectively; then establishing an RIES optimal scheduling model by taking the sum of the power system network loss in the RIES and the cost of purchasing energy from a superior network as an objective function, and giving a power, gas, heat and cold power balance constraint relationship of the RIES; converting the originally built model into a mixed integer second-order cone programming model by using linearization and second-order cone relaxation technologies; calling a Cplex algorithm package on the Matlab provided with the Yalmip optimization toolbox by using the established model according to the predicted operation curves of the fan, the photovoltaic system and various loads, and carrying out optimization solution on the Matlab provided with the Yalmip optimization toolbox; and analyzing the economic benefits generated by SOP and P2G technologies and a lithium bromide absorption refrigerator for reducing the system network loss and cost and improving the wind and light absorption capacity of the system. The invention further comprises a cold and hot gas combined supply dispatching system of the flexible area comprehensive energy system.

The invention discloses a distributed reconfigurable satellite system organization method based on mosaic splicing. The method comprises the following steps: establishing a structural model of a satellite agent; after receiving a task sent by the ground, enabling the satellite agent to quickly establish connection with a nearby satellite agent; splitting the mosaic capability according to the effective load capability of the satellite to obtain mosaic capability modules; performing mosaic splicing on the split mosaic capability modules, and establishing a space mosaic reconfigurable system two-stage system hybrid control structure model with a management satellite agent and a working satellite agent; enabling a management satellite agent to determine a task planning and scheduling scheme of the multi-satellite cooperation, wherein the working satellite agent judges whether an issued task conflicts with an original task of the working satellite agent or not, if yes, the management satellite agent newly formulates a scheme, and the task is executed until the conflict disappears. According to the method, task execution is expanded to multiple dimensions, the task execution effect is comprehensively improved, and the target recognition probability is improved.