54results about How to "Improve schedulability" patented technology

A system and methods for scheduling execution of various types of processes. Using information gathered relating to the different processes, pre-run-time scheduling is integrated of run-time scheduling to guarantee that the executions of the processes will satisfy all the specified relations and constraints. Whenever a new set of processes arrives in the system, the system schedules their execution in two phases: a pre-run-time (off-line) phase performed by a pre-run-time scheduler, and a run-time scheduler (on-line) phase performed by a run-time scheduler.

The invention discloses a multi-imaging satellite emergency task dynamic scheduling method. In order to solve the technical problem, the multi-imaging satellite emergency task dynamic scheduling method is invented to realize a scheduling target on the premise of satisfying the limitation. According to the technical scheme, the multi-imaging satellite emergency task dynamic scheduling method comprises the steps of: synthesizing adjacent target unit tasks located in one field range of a satellite sensor based on task classification; if the allocation of the synthesized tasks is failed, taking out the unit task with the maximum priority in the synthesized tasks through a restoration way for synthesis to therefore decompose the task into a synthesized task and an independent task with more imaging chances; meanwhile, establishing an available time slot in a new task insertion process; moving tasks in a waiting queue backwards; and obtaining a time slot satisfying with the implementation of newly inserted task when the waiting tasks are executable, thereby further improving a scheduling efficiency. By the adoption of the multi-imaging satellite emergency task dynamic scheduling method, tasks can be synthesized, the performing sequence and the performing time of the tasks are adjusted dynamically, imaging time of the satellite is reduced, and accomplishment rate and schedulability of the tasks are improved.

The present invention discloses a method for business dispatching in a time triggering FC network, and relates to the field of an FC network. The method comprises the following steps of: establishinga network model, calculating a cluster period, and determining the length of a single time slot of each time triggering message; determining the priority of a TT message according to a certain rule; planning a link transmitting the TT message; detecting the schedulability of the TT message; selecting the TT message with the highest priority to arrange the time slot of the TT message; according tothe periodicity and the transmission link of the TT message, arranging all other time slots; arranging the TT message of the next priority, and solving a time slot map meeting conflict-free and periodsending of all the TT messages in all the links; and according to a whole network and total business time slot map, solving the sending and receiving time dispatching tables in each terminal and eachexchange. The method ensures the transmitting and receiving of the message determinacy in the optical fiber channel network so as to meet the demand of real-time message dispatching in the complex application system and allow the upper-layer application system performance to be more determined and reliable.

The invention discloses a solar photovoltaic grid-connected inversion control system based on a field programmable gate array (FPGA), which can control a household photovoltaic inversion system to realize grid-connected generation, predict a generation power of the system and realize real-time communication with a grid dispatching centre. The system comprises a signal detection circuit, a control circuit, a communication circuit, a grid-connected circuit and a driving circuit, wherein the signal detection circuit, the communication circuit, the grid-connected circuit and the driving circuit are connected with the control circuit respectively and operate cooperatively under the control of the control circuit. The system predicts the generation power of the household photovoltaic inversion system by using a power prediction algorithm based on a back propagation (BP) neural network and sends a predicted result to the grid dispatching centre in real time. The grid dispatching centre can manage running modes of photovoltaic generation and a conventional power supply reasonably according to the predicted result, reduce an adverse effect on a grid due to incorrect access of a household photovoltaic generation device and improve the schedulability of the photovoltaic generation.

The invention discloses a hardware real-time operation system. The hardware real-time operation system comprises an ARM (advanced RISC machines) processor and an FPGA (field programmable gate array) which are connected via an FSMC (fuzzy sliding mode control) bus, wherein the FPGA is used for realizing a preemptive scheduler, and managements on tasks, a signal quantity, an oil tank, a message queue, a mutual exclusion signal quantity and an event flag group based on the preemptive scheduler; and the ARM processor is used for executing the tasks and realizing task switchover. By implementing the technical scheme of the invention, the FPGA does not occupy the processing time of the ARM processor because of running independently from the ARM processor, and the saved time is used for executing task programs, thus successfully reducing the system expense of the ARM processor; in particular, the utilization rate of the ARM processor is effectively increased in applications needing rapid task switchover. Therefore, via the operation system realized by hardware, the schedulability and real-time performance of the task set are improved.

A power quality promotion method of a light storage power generation system based on hierarchical control provided by the invention comprises the following steps: a, constructing a light storage combined power generation system, wherein the constructed light storage combined power generation system comprises a hybrid energy storage unit arranged on the direct current side of a distributed photovoltaic power generation system; and b, carrying out active management on power quality through power dispatcher control and local power of the light storage combined power generation system. The distributed light storage combined power generation system can effectively stabilize the fluctuation of photovoltaic output power, and strengthens the schedulability of the light storage combined power generation system; and a harmonic current detecting algorithm based on self-adaptive iteration can effectively detect the harmonic component of local load, and provides guarantee for local compensation of the power quality of the distributed light storage combined power generation system. The invention provides a new method for operation and control of the high-permeability photovoltaic power generation system, can effectively promote the power quality of power grid, and provides powerful guarantee for safe and stable operation of the power grid.

The invention discloses a method for optimizing and controlling an energy storage system with variable progressive step length and variable expected outputting. The method comprises the following steps that a single charging and discharging section serves as the progressive control step length, and an SOC critical limiting value circulating charging and discharging model with variable progressive step length and variable expected outputting is established; multi-objective optimization functions with the target of maximum progressive control step length and the minimum overall expected outputting fluctuation ratio are established; on the premise of the SOC critical limiting value circulating charging and discharging model, solution calculation is carried out on the multi-objective optimization functions of the energy storage system based on the NSGA-II algorithm, and the optimal solutions of the multi-objective optimization functions of the energy storage system of a wind power are determined respectively. The maximum ESS utilization rate can be guaranteed through the SOC critical limiting value circulating charging and discharging model, frequent charging and discharging can be avoided through strict charging and discharging state converting, and the continuity of the ESS performance can be guaranteed powerfully. The stabilizing effect of the control strategy can effectively adapt to ultra-short-time prediction and wind electricity economic dispatch level.

The invention discloses a wind-storage combined system scheduling method and apparatus capable of improving the wind power schedulability. The scheduling method includes: the available scheduling output of a wind-storage combined system is calculated via the current charge state of a storage battery and the wind power prediction output; a scheduling model containing the wind-storage combined system is established, and the scheduling model is formed by a target function and restriction conditions; the target function serves as the bacteria fitness value, an improved bacteria foraging algorithm is employed to solve the scheduling model; and if the solution result satisfies all the restriction conditions, the solution result is the optimal solution and is outputted. The scheduling apparatus comprises a calculation module, an establishing module, a solution module, and an output module. According to the method and the apparatus, based on an operation strategy of the wind-storage combined system, the scheduling model of the wind-storage combined system is established, the improved bacteria foraging algorithm is improved for solution, the economy of the operation of the wind-storage combined system is fully considered, the wind power schedulability is effectively improved, the wind power utilization rate is increased, and the method and the apparatus are applicable to the coordination scheduling of the wind-storage combined system.

The invention discloses a method for optimized dispatching of economic benefit of a photovoltaic power generation system. The method comprises the steps of determining a dispatching strategy of the photovoltaic power generation system; building an optimization strategy of the photovoltaic power generation system, wherein the optimization strategy is used for optimizing according to a first allocation proportion when power generation quantity of the photovoltaic power generation system is larger than residual electric quantity of a power utilization load and is used for optimizing according toa second allocation proportion when the power generation quantity of the photovoltaic power generation system is smaller than insufficient electric quantity of the power utilization load; building anoptimized dispatching target function taking benefit of the photovoltaic power generation system within a set period as an optimization target and taking the first allocation proportion and the secondallocation proportion as decision variables; and optimizing the first allocation proportion and the second allocation proportion by employing the dispatching strategy so as to achieve maximum user benefit of the photovoltaic power generation system.

The invention discloses a Deadline-oriented scheduling method capable of meeting time determinability, and belongs to the technical field of airborne network communication. According to the method disclosed by the invention, the scheduling priority of a data frame which is closer to Deadline is higher; meanwhile, scheduling among all ports of a switch is achieved by a fair polling mode; fair service of different links is ensured; the determinability of network exchange delay is ensured. According to the Deadline-oriented scheduling method, the problems that a safe key business requires high real-time scheduling service, and end-to-end delay has an upper bound are solved; the requirements of the delay determinability of an airborne network is ensured to the maximal extent; the data frame which cannot meet the delay requirements can be discarded in advance; the time integrity of the data frame is ensured.

The invention discloses a method and device for trusted data interaction between a block chain and an out-of-chain system, and the method comprises the steps of converting a corresponding uplink data request into a task when a user contract sends a data request which needs to obtain out-of-chain business data, and transmitting the task to a server; when the user contract is in service association with the out-of-chain service system in the execution process, converting a corresponding downlink service driving request into a task, and dynamically acquiring data needing to be interacted through the task. According to the execution mode based on task driving, the flexibility and schedulability of system execution are greatly improved, the operation burden of the node in a complex application scene can be effectively relieved, and the purpose of uplink and downlink reliable data interaction between the block chain and an out-of-chain system can be achieved.

The invention discloses an energy storage system optimization method with wind power prediction errors compensated, which comprises the following steps: (1) the wind power prediction errors of a windpower plant are analyzed to obtain a prediction error distribution law; (2) a mathematical model of rated power and rated capacity of the energy storage system is built; (3) under a certain error compensation degree, an objective function with the purpose of maximizing the income of the wind power plant is built; and (4) the influence of the change of the electricity price and the energy storage cost on the income of the wind power plant is considered from the perspective, and an electricity price interval and an energy storage cost interval which enable energy storage to have investment valueare found out.

The invention discloses a fault-tolerant task-scheduling method based on backward movement in the cloud. The fault-tolerant task-scheduling method is characterized by comprising following steps: acquiring reached information on dependent task groups and information on physical host of a virtualized cloud; utilizing a PB model as a primary edition and a secondary edition set up by each task in dependent task groups; assigning earliest start time and latest finish time for each edition of each task in each dependent task group; marking off multiple virtual machines on each activated physical host and acquiring information on each physical host of each activated physical host; loading each edition of each task in each dependent task group to each virtual machine of each activated physical host within the assigned period of time; operating each edition of each task in each dependent task group on a schedule according to assigned time; and utilizing the method of moving tasks backwards to adjust earliest start time of scheduled tasks as a time window for providing tasks in newly-arriving dependent task groups.

The invention relates to a single-core system real-time task scheduling method aiming at imprecise calculation, comprising the following steps: step 1, establishing a task model according to the taskcharacteristics in the single-core system; step 2, setting up a task model according to the task characteristics in the single-core system. 2, construct an off-line scheduling module and an on-line schedule module according to a task scheduling time point in that task model; 3, construct a task calculation precision unit according to that execution precision of the task for the on-line schedulingmodule; 4, construct an average degree unit for calculating that task exceed the deadline according to the average degree of the task exceeding the deadline for the on-line scheduling module; 5, construct a task exceeding cut-off time frequency calculation unit accord to that frequency of the task exceeding the cut-off time for the on-line scheduling module; The invention guarantees the calculation precision, the cut-off time exceeds the degree and the cut-off time exceeds the upper limit of probability respectively, and on the premise of guaranteeing a certain QoS (Quality of Service), more fully utilizes the systematic calculation resources.

The invention provides a probabilistic real-time task scheduling method based on pWCET (probabilistic worst case execution time) shaping, and belongs to the technical field of real-time systems. A probabilistic real-time task scheduling scheme based on the pWCET shaping is proposed, so that the goals of properly isolating the influence between probabilistic real-time tasks and lowering the calculation complexity of system schedulability analysis are achieved. According to the method, the pWCET of the probabilistic real-time tasks is shaped, and budget allocation is performed on the basis of the shaped pWCET, so that while the probabilistic real-time requirements of the tasks are met, the influence between the probabilistic real-time tasks can be properly isolated, and the schedulability ofthe whole task set is remarkably improved. In addition, the value of the shaped pWCET is obviously reduced compared with that of original pWCET; the calculation complexity of the schedulability analysis can be lowered; and the efficiency is remarkably improved.