63 results about "Dynamic system model" patented technology

The invention provides a carbon emission calculation method based on adaptive Kalman filtering. According to the carbon emission calculation method based on adaptive Kalman filtering, a data matrix is generated by acquiring multi-source data, and a unified data basis is provided for subsequent calculation; a dynamic system model and an error measurement matrix are initialized, a state transition model indicates a carbon emission evolution law, an observation model establishes a mapping relation between multi-source data and carbon emission, and the error measurement matrix dynamically indicates a model effect and adaptively adjusts calculation parameters to ensure the adaptability of the model to the dynamic change of the system; a carbon emission prediction parameter and an error measurement matrix are calculated in real time through adaptive Kalman filtering, a prediction value is corrected through an observation parameter, an optimal estimation parameter and an observation residual error are obtained, meanwhile, the error measurement matrix is updated in real time, and dynamic fusion and noise suppression of multi-source data are achieved through the closed-loop process. And the precision and the real-time performance of carbon emission calculation are improved.

The invention discloses a permanent magnet synchronous motor full-speed-domain position-sensorless hybrid control method and a permanent magnet synchronous motor full-speed-domain position-sensorless hybrid control system, and the permanent magnet synchronous motor full-speed-domain position-sensorless hybrid control method is constructed by taking a permanent magnet synchronous motor nonlinear dynamic system model as a starting point. A hybrid control method combining high-frequency pulsating square wave signal injection under a linear dynamic system model, a rotor polarity judgment and rotor initial position estimation algorithm and an unscented Kalman filtering extension algorithm in an additive noise form under a nonlinear discrete dynamic system model is adopted; and obtaining the optimal estimation observation value of the system state variable in the static, low-speed, medium-speed and high-speed full-speed domain range under the condition that the permanent magnet synchronous motor is not provided with a mechanical position sensor. The optimal estimation observation value is used for compensating and feeding back a permanent magnet synchronous motor drive control system adjusting strategy, disturbance of load torque is improved, the noise amount and disturbance amount introduced by three-phase current measurement are eliminated, and efficient and accurate control operation of the permanent magnet synchronous motor is achieved.

The invention provides a water quality intelligent monitoring method and system, and the method comprises the steps: obtaining the historical water quality data of a water transmission and distribution pipe network, and obtaining parameters in a water quality dynamic system model through a least square method based on the historical data; a parameter estimation result is applied in a laboratory simulation environment, if the difference between model output and measured data does not exceed a preset threshold value, parameters are deployed in an actual water supply system, and the model is dynamically calibrated according to real-time operation data of a water transmission and distribution pipe network; and acquiring real-time water quality data through an Internet of Things sensor, inputting the real-time water quality data into the water quality dynamic system model to predict future water quality indexes, and adjusting water treatment process parameters in advance according to a prediction result. Through the method and the corresponding system, a water quality supervisor can be helped to make an adjustment decision in advance before the water quality is deteriorated, and more intelligent monitoring is realized.

The invention relates to the field of business management, and discloses a business management method based on a station inspection platform, which comprises the following steps: collecting order demands and external environment data, establishing a dynamic system model of the order demands, carrying out Kalman filtering prediction, constructing an optimization model based on an optimal control theory and a variational method, and finally realizing optimization of a resource scheduling strategy. Demand fluctuation and uncertainty are processed, and resource configuration is dynamically adjusted by adopting a multi-level feedback mechanism; the invention further provides a service management system based on the station inspection platform. The service management system comprises a data acquisition module, a nonlinear dynamic model module, a demand prediction module, a resource scheduling optimization module, a random control module and a feedback control module. According to the method, a nonlinear dynamic system model and a Kalman filtering technology are combined, so that accurate order demand prediction and dynamic adjustment are realized; resource scheduling is optimized based on an optimal control theory, a random control model is introduced to cope with uncertainty, and a multilevel feedback mechanism ensures quick response and adjustment.

The invention provides a mechanical arm imitation learning method and device based on DMP, and belongs to the technical field of robots. The method comprises the steps that a trajectory is fitted through a motion primitive method; according to the teaching or simulation track, the weight of the motion primitive is obtained through a least square method; a new trajectory regeneration method is provided, and new trajectories with the same geometric features are generated by setting a new starting point, a new ending point or a new scaling coefficient; a new track passing point strategy is provided, so that the generated new track can pass through a specific key point; applying a Gaussian process to generate a high-order smooth trajectory passing through a specific point; generating a high-order smooth curve meeting the geometric characteristics by using a weighted average method; substituting the generated new trajectory into the new DMP model, and establishing a dynamic system model; and according to the dynamic system model, the mechanical arm joints are driven through the joint speed track, and speed control is achieved. According to the method, a large amount of similar industrial programming work can be reduced, and the flexibility and efficiency of trajectory planning and control are improved.

Methods, systems, and apparatuses for combining pressure measurement with a dynamic system model to estimate the air mass within an actuator, and then estimate the applied contact force which may be based on a quasi-static deformation model.

The invention relates to the technical field of data processing, and discloses a nuclear radiation sensor data processing method based on Kalman filtering and an alarm apparatus, and the method comprises the steps: continuously obtaining nuclear radiation observation data according to a preset sampling period; acquiring Kalman filtering observation data at a previous sampling moment, and predicting Kalman filtering observation data at a current sampling moment by using a preset Kalman filtering model according to the Kalman filtering observation data at the previous sampling moment to obtain a current prediction result; and correcting the current prediction result by using the observation data at the current sampling moment to obtain Kalman filtering observation data at the current sampling moment, and displaying the Kalman filtering observation data. According to the method, the dynamic system model is established based on the Kalman filter, the state at the current sampling moment is predicted based on the Kalman filtering observation data at the previous sampling moment, the evolution of the system state is accurately estimated in combination with the measurement data of the sensor, the dynamic change of the system is better coped with, and the accuracy and reliability of nuclear radiation detection are improved.

The invention discloses a dynamic load response adjustment method and system for a hot water boiler, and relates to the field of boiler load adjustment, and the method comprises the steps: carrying out the deep learning of boiler and weather data through LSTM, and precisely predicting the thermal load change trend in a period of time in the future; on the basis, the current system state of the boiler is estimated in real time in combination with a Kalman filter, and the predicted load, the real-time state and a dynamic system model of the boiler are incorporated into a model predictive control (MPC) framework. And the MPC calculates an optimal gas valve opening sequence considering the physical time lag of the boiler through rolling optimization in a prediction time domain. The prospective prediction is combined with optimization control based on model dynamic characteristics, so that the boiler can adjust fuel supply in advance, response delay caused by system thermal inertia is effectively overcome, rapid and accurate response to dynamic loads is achieved, and the defect that traditional control is always slow by half beat is overcome.

The invention relates to the technical field of medical image processing, in particular to a cardiovascular surgery patient circulation state image real-time monitoring and early warning system. Comprising a video acquisition module, an image preprocessing module, a microcirculation evaluation module, a multi-part perfusion fusion module based on differential manifold, a trend prediction and early warning module and a clinical interaction and data integration module. A local perfusion index and a pulsation stability index are extracted after preprocessing, multi-part microcirculation signals are creatively mapped to a differential manifold for characterization, time-frequency dual-channel fusion driven by Riemannian geometry and an attention weighting mechanism are integrated into a whole-body perfusion comprehensive index, modeling is carried out based on a dynamic system on the manifold, and a whole-body perfusion model is established. Early warning can be provided before the conventional blood pressure is obviously reduced, precious time is gained for clinical intervention, invasive complications are avoided through a non-invasive monitoring mode, and whole-course continuous monitoring can be achieved.

The application discloses a method for adaptive secure consensus control of a multi-agent spatio-temporal dynamic system with unknown boundary nonlinearity under hybrid attacks, comprising the following steps: constructing a multi-agent spatio-temporal dynamic system model with unknown boundary nonlinearity and a virtual leader model; establishing a hybrid network attack model containing deception attacks and denial of service (Dos) attacks, and using an adaptive radial basis neural network to approximate the unknown boundary nonlinearity function; defining a consensus error signal to obtain a consensus state error system; designing a composite adaptive neural network secure boundary consensus control scheme, constructing a Lyapunov function for the error system, and obtaining a sufficient condition for the error system to realize mean square secure consensus control. The application effectively solves the problem that the traditional method is difficult to simultaneously cope with complex network attacks and unknown boundary nonlinear disturbances, and significantly improves the security and robustness of the system.

The application provides a response lag compensation method of a urea liquid injection system, which solves the problem of response lag between the urea liquid rate supplied by a metering pump and the urea liquid rate sprayed by a gas-assisted nozzle in the gas-assisted urea injection system, and comprises the following steps: a first-order dynamic system model about the residual urea liquid amount in a urea pipe is constructed to associate the urea liquid rate sprayed by the gas-assisted atomizing nozzle with the urea liquid rate pumped by the metering pump; an imbalance difference between the urea liquid rate pumped by the metering pump and the urea liquid rate sprayed by the gas-assisted atomizing nozzle is obtained, the urea liquid rate sprayed by the gas-assisted atomizing nozzle at t2 is calculated as X2, the required urea liquid injection rate at the next moment t2 is given as Xneed according to the actual working condition of the engine, the urea liquid rate pumped by the metering pump at the next moment t2 is calculated as Y1, the urea liquid rate pumped by the metering pump at the next moment t2 is corrected as Y2 which is calculated to compensate, and in addition, the application also provides a gas-assisted urea liquid injection system applying the method.

The invention discloses a robot teaching obstacle avoidance method based on a neural network and a dynamic system model. The method comprises the following steps: acquiring a demonstration data set; coding the current robot state into a high-dimensional feature vector; nonlinear feature components are extracted from the encoded features, weighted summation is carried out on the features, quadratic terms are added, and an energy function is constructed; constructing a decision function, and multiplying the state and environment information of the robot by corresponding weights to obtain the output of the decision function; inputting the output of the decision function into a multi-task hybrid neural network, generating a parameter vector, constructing a lower triangular matrix based on the parameter vector, and constructing a modulation matrix through cholosky decomposition; constructing a loss function, and obtaining an optimal model parameter through a nonlinear programming method; and taking the output of the neural dynamic system model as a robot control strategy to complete a specified task. According to the invention, by introducing the dynamic system model and the neural network, the bottleneck problems of time dependence and dynamic obstacle avoidance are solved.

The present invention relates to a method, apparatus, and storage medium for intelligently adjusting the size of a network card receive ring. The current bandwidth utilization rate, current latency, current packet loss rate, and current network card receive ring size for one cycle are obtained. Based on a non-linear dynamic system model, the network card receive ring size for the next cycle is predicted according to an adaptive adjustment factor, the current bandwidth utilization rate, the current latency, the current packet loss rate, and the current network card receive ring size, resulting in a predicted network card receive ring size. The adaptive adjustment factor is adjusted according to the feedback error and a non-linear function, and the parameters of the non-linear dynamic system model are adjusted according to a loss function. The size of the network card receive ring is adjusted based on the predicted network card receive ring size. According to the current bandwidth utilization rate, the current latency, the current packet loss rate, and the current network card receive ring size, the size of the network card receive ring is intelligently adjusted, and by optimizing the adjustment of the adaptive adjustment factor and the parameters of the non-linear dynamic system model, the accuracy of the network card receive ring test results is improved.

The invention discloses a power grid new energy bearing capacity measurement and calculation method, and relates to the technical field of power grid bearing capacity measurement and calculation, and the method comprises the following steps: constructing a power grid dynamic model comprising a power generation unit, a power transmission and transformation network, a load node and a new energy access point; establishing a basic safety margin boundary, and generating a static power flow model; constructing a dynamic stability model reflecting the influence of new energy fluctuation on power grid frequency and voltage dynamic response by using a dynamic simulation platform; and dynamic stability analysis: new energy inverter access points are added in the power grid dynamic model to construct a dynamic system model, a virtual inertia control module is embedded in the new energy inverter model, and a control loop is constructed. According to the method, a final control instruction is generated by constructing a static model and a dynamic model, embedding a virtual inertia control module in an inverter model, constructing a control loop and combining virtual inertia compensation power with inverter original power.

The present application relates to the technical field of medical image processing, in particular to a real-time monitoring and early warning system for the circulation state image of a cardiovascular surgery patient, comprising a video acquisition module, an image preprocessing module, a microcirculation evaluation module, a multi-site perfusion fusion module based on differential manifold, a trend prediction and early warning module, and a clinical interaction and data integration module; the video of the patient's fingernail bed and lips is simultaneously acquired by a high-frame-rate camera, and after preprocessing, the local perfusion index and pulsatility stability index are extracted; innovatively, the multi-site microcirculation signals are mapped onto the differential manifold for representation; through the time-frequency dual-channel fusion driven by Riemann geometry and the attention weighting mechanism, the whole-body perfusion comprehensive index is integrated; and based on the dynamic system modeling on the manifold, early warning can be provided before the conventional blood pressure significantly decreases, valuable time is won for clinical intervention, the non-invasive monitoring method avoids invasive complications, and whole-process continuous monitoring can be realized.

The application relates to a multi-agent cooperative control method and device for transmission data loss, and belongs to the technical field of multi-agent control. In the method, a directed network topology thought is introduced into information interaction of a multi-agent system, a Bernoulli distribution is introduced to express packet loss in a data transmission process between agents, a second-order dynamic system model of the agent is constructed to model the state, a following error equation of the multi-agent system is established, Liapunov stability theory and robust control theory are used to analyze the tracking error equation, a stability control condition is obtained, and thus a control gain coefficient capable of realizing consistent stability of the multi-agent system can be obtained by solving the stability control condition.

This invention belongs to the field of measurement and control, and relates to a fixed-time preset performance control method, device, computer equipment, and storage medium for a macro-micro composite motion platform. The method includes: constructing a dynamic system model of the macro-micro composite motion platform; transforming the dynamic system model into a state-space equation suitable for a backstepping recursive control framework; designing a fixed-time preset performance function to constrain the tracking error signal within a preset range within a fixed time; constructing a nonlinear filter to simplify the control design process and establishing a switching function to resolve coupling terms in the system; designing an adaptive fixed-time preset performance tracking control algorithm and proving the stability of the closed-loop system based on Lyapunov stability theory. By designing the controller using an adaptive backstepping recursive control framework, the control design process is simplified, and the computational load is reduced; the tracking error signal of the closed-loop system converges to the preset range within a fixed time, improving the transient performance and control accuracy of the system.

A method of monitoring an elevator car in an elevator shaft includes acquiring position data indicative of a position of the elevator car, acquiring motion data indicative of a motion of the elevator car, and determining, from a dynamical system model, an estimated position of the elevator car. The dynamical system model describes a motion of the elevator car based on input variables that include the position data and the motion data. The method further includes determining an offset value indicative of a motion data offset, wherein the offset value is generated such that the dynamical system model fits the position of the elevator car indicated by the position data, determining a sensor reliability parameter based on the offset value, and providing output data including the sensor reliability parameter.

This invention belongs to the technical field of fuel cell gas supply systems and provides a control method and system for PEMFC injection gas supply systems based on deep reinforcement learning. Addressing the difficulty in establishing accurate control models for PEMFC injection gas supply systems, this invention proposes a control method for PEMFC injection gas supply systems based on deep reinforcement learning. First, a dynamic system model of the PEMFC injection gas supply system is established using a deep neural network. Second, an actor-critic framework is used to interact with the learned dynamic system model of the PEMFC injection gas supply system and maximize the cumulative reward within the prediction interval, thus learning a neural network strategy based on model predictive control. Finally, by fixing the parameters of the actor network model and deploying the actor network in the controller of the PEMFC injection gas supply system, real-time optimal control of the PEMFC injection gas supply system can be achieved.

The application discloses a random complex network state estimation method and system, and belongs to the technical field of estimators. When the measurement signal transmission requirement of a target node obtained in real time meets a preset dynamic event triggering condition, an augmented error system model is constructed based on a constructed dynamic system model, a missing measurement model and the dynamic event triggering condition, and a gain matrix of a constructed state estimator is obtained by solving a linear matrix inequality. The state estimator is updated according to the gain matrix, and the measurement signal of the target node at a triggering moment is input into the updated state estimator to obtain an updated state estimation signal at a current moment. The triggering condition includes a preset waiting time interval and a forced triggering time constraint. The internal dynamic variable of the dynamic event triggering condition is updated according to the updated state estimation signal at the current moment. The method balances the state estimation precision and the communication frequency by introducing a dynamic event triggering mechanism.

The invention discloses a multi-sensor information fusion estimation method for state-dependent observation loss, and the method comprises the steps: constructing a system model which is a linear discrete dynamic system model and is used for representing a dynamically changing system state; training an observation loss discriminator, processing a complex observation rejection domain by the observation loss discriminator through a support vector machine, and converting a nonlinear classification problem into a linear convex optimization problem through a kernel function so as to discriminate whether a target state is in the observation rejection domain; and constructing a recursive fusion estimator based on an EM algorithm. According to the invention, the recursive estimator has high robustness and adaptability, can effectively deal with the complex situation of state dependence observation loss in various actual scenes, and significantly improves the precision of system state estimation.

Methods, systems, and apparatuses for combining pressure measurement with a dynamic system model to estimate the air mass within an actuator, and then estimate the applied contact force which may be based on a quasi-static deformation model. An example method includes, in response to a request to apply a contact force by an actuator at a first actuator pressure, determining a second actuator pressure for the actuator to apply the contact force. The contact force is generated using one or more components of a haptic assembly, and the second actuator pressure is determined based on one or more of properties of the components of the haptic assembly and pressure data obtained via one or more sensors. The example method further includes causing the haptic assembly to adjust the first actuator pressure to the second actuator pressure such that the actuator applies the contact force.

The invention discloses a preset performance control method based on a normalized error boundary function, and relates to the field of automatic control. The method comprises the following steps: establishing a dynamic system model; selecting a preset performance function based on the kinetic model, carrying out normalized error boundary function design based on the selected preset performance function, and solving error conversion based on the normalized error boundary function and the conversion function; and introducing a conversion error, designing a nonsingular terminal sliding mode surface and a reaching law, and completing the design of the preset performance controller based on the normalized error boundary function. Compared with the prior art, the control method has the advantages that the system state can track the state instruction more efficiently, the control method is prevented from losing efficacy when the error exceeds the preset performance function boundary, and the transient performance and the steady-state performance of the control system are improved.

This invention discloses a robot navigation method suitable for large sparse environments, comprising the following steps: Step 1, establishing a system evolution equation based on the robot's state vector and kinematic equations, incorporating an environmental sparsity factor into the system evolution equations, and then introducing trajectory tracking weights and obstacle avoidance weights to construct a dynamic system model; Step 2, determining whether the dynamic system model is in a critical state between stability and chaos by solving the Jacobian matrix and Lyapunov exponent of the dynamic system model. When the maximum Lyapunov exponent is approximately equal to 0, the critical stability condition is reached; Step 3, based on the critical state, designing a path evaluation function to screen stable path points, and dynamically adjusting the obstacle avoidance weights through bifurcation parameter adjustment control equations to achieve adaptive switching between ordered tracking and chaotic obstacle avoidance, thus completing robot navigation in large sparse environments. This invention can ensure that robots can efficiently and stably complete navigation in large sparse environments.

The invention relates to a dynamic system model predictive control method based on RBF neural network, which belongs to the field of control technology of dynamic system. The invention solves the problems of poor control accuracy and slow dynamic response speed in the existing dynamic system model predictive control method. In the invention, a dynamic equation between the control input and the state variable of the dynamic system is established according to the dynamic parameters of the dynamic system, and the dynamic equation is converted into a nominal form based on the virtual control variable by using a virtual control variable, and then converted into a state space form. The cost function of the model predictive control algorithm is designed by using the dynamic equation in the state space form, and the cost function is optimized and solved. The disturbance approximation model based on the neural network is adopted, and the unmodeled dynamics and external disturbances of the dynamic system are obtained by inverse solution. The unmodeled dynamics and external disturbances are used to obtain the system output after disturbance compensation, and then the control signal of the dynamic system is obtained. The invention is suitable for the dynamic system control of quadrotor aircraft.

The invention discloses a four-rotor unmanned aerial vehicle obstacle avoidance tracking control method and system under a limited environment, and belongs to the technical field of industrial process control. The method comprises the following steps: constructing a four-rotor unmanned aerial vehicle dynamic system model; establishing a state environment limited model according to the state error; estimating the nonlinear state of the quad-rotor unmanned aerial vehicle; a command filter and a virtual control rate model are constructed, and the problem of derivation complexity explosion is avoided; and constructing a four-rotor unmanned aerial vehicle state controller model, and performing real-time control on the state of the four-rotor unmanned aerial vehicle. According to the invention, environment limited control is introduced, so that unmanned aerial vehicle control is realized efficiently and smoothly under the condition that constraints are applied to different external conditions or hardware physical limitation changes; meanwhile, a potential energy function is introduced under the condition that the state is limited, and collision and obstacle avoidance under the condition that the environment is limited is achieved.

The invention provides a multi-unmanned aerial vehicle wireless energy transmission and wireless data collection collaborative optimization method based on adaptive hierarchical deep reinforcement learning, comprising: constructing a dynamic system model composed of a plurality of unmanned aerial vehicles, a plurality of information devices and a plurality of energy devices, the information devices uploading data to the unmanned aerial vehicles, and the energy devices uploading data to the unmanned aerial vehicles; the energy equipment receives wireless energy from the unmanned aerial vehicle; minimizing the total information age of all the information devices and minimizing the total energy hunger degree of all the energy devices are set as common optimization objectives, and the double-objective optimization problems are fused into a single-objective optimization function according to weights; and constructing and operating a three-layer multi-agent deep reinforcement learning framework, solving the single-target optimization function, and outputting an optimal cooperative control strategy. According to the method, real-time balance, autonomous decision making and efficient cooperation of the multiple unmanned aerial vehicles for heterogeneous Internet of Things equipment requirements are realized.

A horticulture growing system where a growing regimen is prescribed to achieve desired growing results. The system has self-learning mechanisms where the prescribed growing regimens are continually optimized to achieve the desired results through machine learning and deep learning. The system uses both a cloud based dynamic system model for growing and a local grow model. Various techniques are utilized to improve data collection and labeling. The results (the system's ability to accurately create growing regimens which produce the desired grow objectives) are improved using the dynamic system model and the local grow model. The models are trained and adjusted using datasets from multiple growing operations to increase the efficacy of the self-learning mechanisms. This system may also include a mechanism for in-harvest re-optimization to improve grow results in real-time.

The present invention discloses a method for quantifying dynamic measurement uncertainty based on physical information and evidence learning, comprising the following steps: Step 1: Obtaining an objective function of the output result of a constrained neural network based on a target dynamic system, and obtaining a training data set and a test data set; Step 2: Based on an evidence theory uncertainty quantification method, constructing and generating four independent feedforward neural networks, training the feedforward neural networks using the training data set, and obtaining a dynamic system model consisting of the four trained feedforward neural networks; Step 3: Inputting the test data set into the dynamic system model, and calculating the total uncertainty of the prediction result based on the tensor of the evidence distribution output by the four trained feedforward neural networks. The present invention improves the generalization ability and physical interpretability of the neural network, and avoids the situation in traditional methods where the final uncertainty change trend is the same as the model input due to the dependence of uncertainty propagation on the model input.