65 results about "Fault tolerant algorithms" patented technology

A fault tolerant control system for a multi-phase permanent magnet assisted synchronous reluctance motor utilizes vector control to provide safe operation under various phase loss fault conditions. Specifically, the vector control of the present invention utilizes a fault tolerant algorithm that receives a torque input and an electrical current feedback signal from the motor. Thus, in the presence of a fault condition, the vector control applies the optimal torque angle to the motor, while reducing the phase currents to an optimized value to lessen the saturation effect in the motor, so as to ensure that the motor delivers maximum torque output in the presence of such faults. As such, the control system allows the motor to operate safely with high reliability, which is highly desirable, such as in electric vehicles and the aerospace industry.

The invention provides a private block chain block building method for a Byzantine fault tolerant algorithm of quartic communication. The method comprises: (1) all nodes carrying out hash mapping on the received transactions to obtain a bitarray sent to all remaining nodes, and each node carrying out 2/3 AND operation on the received bitarray to obtain a bitarray corresponding to a transaction intersection of over 2/3 nodes; (2) a block building node obtaining a transaction set according to the bitarray for building a block, and submitting the block to the remaining nodes; (3) a node receiving the block completing verification by comparing self bitarray with the transaction set in the block, and sending a digital signature of the verification result to all the remaining nodes after verification; and (4) forwarding all vote signatures received by all nodes on the block in second vote so that each node receives votes of all the nodes, and collecting statistics on the votes to obtain a final result, thus determining whether to accept the block.

The invention provides a consensus method, device and system based on effective computational force contribution. The method includes the steps of receiving computational tasks issued by a computational publisher and computational requirements corresponding to the computational tasks; configuring computing tasks according to computing requirements; obtaining the data needed by the computing task and compiling the computing task into a Boolean circuit; forming a plurality of parallel computing tasks, and distributing the parallel computing tasks to a plurality of computing nodes for computing;receiving the calculation result and the calculation proof of the parallel calculation task by the calculation node, and determining whether the calculation result is valid or not according to the calculation proof; if it is valid, determining the cumulative calculation contribution value of the calculation node corresponding to the calculation result, and allocating the preset reward for each calculation node; receiving voting data of each computing node as cumulative calculated contribution value, and voting on the candidate nodes; obtaining the N consensus nodes, wherein the practical Byzantine fault-tolerant algorithm is used to complete the consensus of the block data among the N consensus nodes.

The invention relates to a finite-time posture fault-tolerant control method for a spacecraft formation and belongs to the technical field of multi-spacecraft formation flying. In order to reduce communication paths and avoid resource waste, a directional communication topological structure with small communication traffic is adopted between spacecraft formation members; in addition, in order to make a spacecraft formation posture control system achieve synergy more quickly and stably, the finite-time posture fault-tolerant control method for the spacecraft formation is provided on the basis of a finite-time algorithm with high robustness and fast convergence, a redundant fault-tolerant algorithm with better fault-tolerant performance and an self-adaptive algorithm with a good suppressioneffect on system uncertainty and interference. According to the method, a limited-capacity self-adaptive posture cooperative tracking fault-tolerant control strategy is perfected, a redundant fault-tolerant control method is achieved through reasonable torque distribution, and at the same time, a self-adaptive law is designed for reducing the influence of inertia variation and interference; moreover, fast convergence of a spacecraft cooperative tracking error system can be realized, and the robustness and practicability of the control system are further improved.

The invention provides a data storage method and device based on a redundancy mechanism and a server. At the server side, the method comprises the steps that the state of a data file is checked according to a preset period, and a cold data file with an access rate smaller than a first preset access rate threshold is found; the cold data file is fragmented according to a preset data fragmentation backup fault-tolerant algorithm, and all data fragments are stored into different storage nodes; the cold data file is deleted; and storage node information of all the data fragments is recorded. According to the method, the data file with a low access rate is fragmented and stored by use of the data fragmentation backup fault-tolerant algorithm (e.g., an IDA data fragmentation algorithm), and a certain storage space can be saved on the premise of guaranteeing data security and integrity.

The invention discloses a fault-tolerant control method for a single-phase open-circuit fault of a five-phase permanent magnet synchronous motor with improved SVPWM. The method comprises the followingsteps: obtaining a variety of SVPWM fault-tolerant control methods corresponding to different solutions of control parameters by the algorithm according to the existing five-phase SVPWM fault-tolerant control theory, solving the optimal solution of the control parameters to reconstruct the post-fault system coordinates by analyzing the direct current bus voltage utilization and phase current harmonic suppression ability of the motor, and performing redistribution and district reconstruction of spatial vectors on the reconstructed coordinate system. Compared with the prior art, the fault-tolerant control method effectively reduces the current harmonic content, restrains the motor torque ripple, and simultaneously improves the direct current bus voltage utilization rate while ensuring thatthe average torque outputs of the motor before and after the single-phase open circuit fault are constant. More critically, the fault-tolerant algorithm may achieve high dynamic performance fault-tolerant operation of the motor and has certain versatility.

The invention relates to a spacecraft amplitude limiting adaptive attitude cooperative fault tolerant control method, which belongs to the technical field of multi-spacecraft formation flying. A redundant fault-tolerant algorithm is used to process a spacecraft fault. A control torque saturation controller carries out torque amplitude limiting. An adaptive algorithm control term compensates for inertia variation and external disturbance, and a fast convergence control algorithm is used for fast attitude cooperation. Other algorithms are used. The spacecraft amplitude limiting adaptive attitudecooperative fault tolerant control method is proposed. According to the invention, the consideration of a torque actuator fault, torque amplitude limiting, external disturbance and inertia variationuncertainty is more advanced; a spacecraft cooperative tracking error system can quickly converge; and the robustness and practicability of a control system are further improved.

The invention relates to the field of block chains, in particular to a practical Byzantine fault-tolerant algorithm improvement scheme for an alliance block chain, which is a consensus scheme capable of being used for the alliance block chain. The consensus algorithm mainly used in the alliance block chain is a practical Byzantine fault tolerant algorithm (PBFT), but the PBFT algorithm has the problems of low security, low efficiency and the like. According to the method, the characteristics of the alliance block chain are utilized, and an improved PBFT (Credit Recurrent PBFT, CRPBFT) scheme is provided. In the scheme, a credit mechanism is introduced to monitor the consensus behavior of the node to ensure that the Byzantine node is effectively checked out, so that the probability that the Byzantine node becomes a consensus node is reduced; secondly, nodes are selected according to the resource contribution amount, the reliability and stability of the nodes are guaranteed to the maximum extent, and the system enters a virtuous circle; and finally, the security and consensus efficiency of the algorithm are improved by adding a candidate consensus node set and removing Byzantine nodes.

The invention provides a reconfigurable fault-tolerant type driving topological structure of a redundant power source and belongs to the field of high-reliability motor driving control systems. The reconfigurable fault-tolerant type driving topological structure of the redundant power source comprises double power sources, N H-bridges, fuses, a reconfiguration switch and the like. The reconfigurable fault-tolerant type driving topological structure of the redundant power source is characterized in that in normal work, an independent H-bridge driving topological structure is applied to each phase, reconfiguration of the driving topological structure is achieved through switching on or switching off of the reconfiguration switch on the driving topological structure, namely, a fault can be isolated through the reconfiguration switch when the open-circuit fault or the short-circuit fault occurs on the power sources, power tubes and the reconfiguration switch, according to the type of the fault and the needs, the independent H-bridge driving topological structure of each phase can be reconfigured into star-shaped driving topology, finally, the fault tolerance function is achieved, and the reliability of the motor driving control system is improved. The reconfigurable fault-tolerant type driving topological structure of the redundant power source has the advantages of being flexible in power supply, capable of rapidly and conveniently isolating the fault and conducting hardware reconfiguration, rapid, strong in fault tolerance ability, flexible in use, free of an extra fault-tolerant algorithm, high in reliability and the like.

The invention discloses a sinusoidal or trapezoidal counter potential five-phase permanent magnet motor short-circuit fault-tolerant control method adopting dead-beat current tracking. According to the method, the third harmonic space of the five-phase permanent magnet motor in a fault state is fully utilized, and a fault-tolerant control strategy for the five-phase permanent magnet motor with sine counter electromotive force and trapezoidal counter electromotive force under the condition of a single-phase short-circuit fault is provided, so that the third harmonic space of the five-phase permanent magnet motor can be fully utilized during fault-tolerant operation of the five-phase permanent magnet motor. According to the invention, the performance of the output torque of the motor in a fault state is improved, and the fault-tolerant operation efficiency of the motor is improved, so that the motor can have better fault-tolerant performance and dynamic response; the speed regulation range during fault-tolerant operation of the motor is widened, so that the motor can be better applied to the application fields, such as electric vehicles, requiring high reliability and wide speed regulation range; and compared with a current hysteresis modulation method used in a traditional fault-tolerant algorithm, the switching loss of the inverter is reduced.

Disclosed is a fault-tolerant algorithm of a human body sensor. The human body sensor is provided with a power circuit, a microprocessor, a human body infrared sensor body which is connected with the microprocessor, a microwave Doppler sensor, a switch driving circuit and a control switch which is connected with the switch driving circuit. The fault-tolerant algorithm set by the microprocessor comprises the following steps that (1) under the situation that the control switch is connected, the microprocessor conducts N times of sampling on the human body infrared sensor body in the sampling period T1, and if the number of times when a human body is detected is zero, then the step 2 is executed; (2) the microprocessor turns off the control switch and conducts M times of sampling on the microwave Doppler sensor in the delaying time T2, and if the number of times when a human body is detected is zero, then the step 3 is executed; otherwise, the microprocessor conducts the control switch completely again and the step 1 is executed; (3) the microprocessor turns off the control switch completely, and conducts N times of sampling on the human body infrared sensor body in the sampling period T1, if the number of times when the human body is detected is not zero, the microprocessor conducts the control switch and the step 1 is executed.

The invention discloses a steering-by-wire system test bench, which comprises the design of a master control module, a fault-tolerant module, a steering control module, a steering execution module and a resistance loading module and connection parts thereof. A steering wheel is connected with a steering-column-assisted type steering gear to simulate road surface feedback; and two pinion-and-rack type steering gears with different power are connected by means of the parts, receive sensor signals by means of a master control single chip microcomputer, and realize steering and resistance simulation through transmitting PWM signals by using an Flexray bus. The mechanical connection between steering execution and operation is reduced, a transmission ratio and road feeling can be designed freely to change force transmission characteristic and angle transmission characteristic of the system, the layout space of a vehicle is saved, and the operation stability of the vehicle is improved. The steering-by-wire system test bench can simulate the road surface resistance in various states, and improves the precision of the system. The construction method of the SBW test bench disclosed by the invention is simple and effective, and can be applied to the experimental study of an SBW system control algorithm, a fault-tolerant algorithm and a bus technology.

The invention relates to a block chain medical data sharing excitation method based on a Shapley value, and belongs to the technical field of block chain network application. A block chain medical data sharing excitation method is established; alliance revenue is distributed among cooperation alliance members by using a Shapley value method, the excitation problem of medical data sharing in a block chain application scene is solved, and the proposed block chain medical data sharing excitation method can effectively excite the cooperation alliance members to participate in cooperation and sharereal data. According to the method, the contribution degree of a data owner and the weights of all members in a cooperation alliance are considered, so that a revenue distribution scheme based on a Shapley value method is suitable for a medical data sharing scene with contribution degree difference; a fair reward mechanism is formulated for consensus nodes in an alliance chain network according to practical Bayer's space fault-tolerant algorithm characteristics applied to the alliance chain.

The invention relates to the field of a consensus algorithm of a block chain distributed node, in particular to a hierarchical concurrent Byzantine consensus method and system applying a digital signature technology. The method comprises the following steps: a node which creates a block in a sliding time window is elected as a leader node to participate in consensuses, a creation node of a newestblock is used as a supreme leader node, and a stock right pyramid is built from big to small according to stock rights; the stock right pyramid is sequentially checked and adjusted from top to bottom;after the adjustment is completed, the consensus of each piece of transaction information is initiated by the supreme leader node, and the consensuses are carried out in each consensus group sequentially from top to bottom; message transmission is carried out in a digital signature mode during the consensuses; after the consensuses of all consensus groups are completed, a consensus result of current transaction information is broadcast to the whole network, and then the consensus of the next piece of transaction information is started; and when a new block is created, the sliding time windowslides backwards for one block, and the above steps are repeated. The defects of an application of a Byzantine fault-tolerant algorithm to an open block chain can be overcome.

The invention designs an optimized practical byzantine fault tolerance (OPBFT)applicable to a block chain. The OPBFT is used for solve the problems that a traditional Byzantine algorithm cannot dynamically sense the number of nodes and the efficiency of electing malicious nodes is low. In order to solve the problem of node adding, a node dynamic adding and quitting mechanism is introduced in the method, node adding and quitting need mutual confirmation among all nodes, and the confirmation process is divided into four stages, namely request, confirmation, reply and completion. In order to solve the problem of randomness of main node selection, a scoring and voting collaborative mechanism is introduced, the scoring mechanism follows the longest chain principle of a block chain, and the voting scoring mechanism is cooperatively carried out. A node adding and quitting mechanism in the method enables the number of consensus nodes to be variable, and the practicability of a consensus algorithm is greatly improved. A scoring and voting cooperation mechanism greatly ensures the correctness of the selected main node and improves the security of the system.

The invention relates to an alliance blockchain network consensus method based on mixed Byzantine fault tolerance, and belongs to the technical field of blockchains. The method comprises the steps that S1, a fragmentation mechanism is utilized to classify blockchain network nodes into four node types including common nodes, verification nodes, sorting nodes and block output nodes, wherein the output blocks are all nodes; S2, due to the characteristics of a parallel Byzantine fault-tolerant algorithm, dividing a consensus mechanism into two stages, including a preprocessing stage algorithm andan execution stage algorithm, and executing the preprocessing algorithm; S3, in order to ensure the correctness of the parallel Byzantine fault-tolerant algorithm, the normal operation of the main node in a consensus mechanism is ensured by utilizing a main node view switching algorithm; and S4, the parallel Byzantine fault-tolerant algorithm is executed by the output block nodes to achieve a consensus of the output block nodes, and block output is completed by the main node in the consensus of this round. Compared with the consensus mechanism in the current alliance block chain, the securityof the consensus mechanism and the consensus reaching efficiency of the node are improved.

The invention discloses a distributed flight control computer control system based on MPC8280. The MPC8280 and a FlexRay bus are effectively combined and used for the aviation field. The system meets the requirements of flight control, flight simulation and other applications, makes up for the blank that the MPC8280 and the FlexRay bus are used for the field of flight control in the prior art, and solves the problem that a fault tolerant algorithm, an aircraft control law and task logical processing are complex in the existing field of flight control.

The invention discloses a crowdsourcing transaction method and system, a readable storage medium and a terminal. The method comprises the following steps of receiving the registration information of apacket receiver and a packet sender through intelligent management equipment in a crowdsourcing transaction platform, and when the registration information of the packet sender and the packet receiver passes the authentication, writing the registration information of the packet sender and the packet receiver into the block chain in the crowdsourcing transaction platform by adopting an improved Bayer fault-tolerant algorithm. According to the scheme, the malicious nodes can be effectively prevented from tampering data, the problem of leakage of crowdsourcing transaction data is avoided, and therefore the security of crowdsourcing transaction can be improved.

The invention discloses a block chain backup for disaster recovery system, method, server and a computer readable storage medium. Block chain nodes are deployed to a first primary machine room, a standby machine room and at least one second primary machine room. When the downtime node amount is greater than the fault-tolerant capacity theoretic value of the practical Byzantine fault-tolerant algorithm, the standby machine room is started through the block chain backup for disaster recovery server to perform the backup for disaster recovery. Compared with the prior art, the system availabilityof the system is guaranteed while performing the backup for disaster recovery of the block chain.

The invention discloses a software-defined opportunistic network flow table updating method based on a practical Byzantium fault-tolerant algorithm. The method comprises the following steps: firstly,a control layer sending an initial flow table to be updated to all switches; the first switch forwarding the received initial flow table to be updated; when the other switches receive the first flow table forwarded by the first switch, the first flow table being switched to the second flow table; matching the content of the first flow table with the content received by the first flow table; if thematching is successful, the first flow table being continuously forwarded to other switches; and if the number of the first feedback information received by a certain switch exceeds a preset value, the switch entering a confirmation state and forwards own states to other switches, and when the number of the switches finally entering the confirmation state exceeds the preset value, each switch updating the flow table. According to the method, the PBFT algorithm of the block chain is used for solving the problem of inconsistent updating of the flow table, reliable distribution of the flow tablecan be ensured, and extremely high fault tolerance is achieved.