312results about How to "Improve recovery effect" patented technology

The invention discloses a low-orbit communication satellite load multilink cooperative transmission system, and belongs to the technical field of satellite communication. In order to solve the problems of insufficient link reliability, low utilization rate of resources in constellations, weaker emergency response and fault-tolerant capability and higher operation cost, the invention is composed of a plurality of low-orbit communication satellites, realizes global seamless coverage through reasonably deployed constellation configuration, ensures that a user can obtain stable communication service in any area, and improves the user experience. A multi-link cooperative communication mechanism is adopted, through real-time link quality evaluation and dynamic link selection, rapid switching of link faults is achieved, communication reliability is improved, a dynamic path adjustment technology is combined, a forwarding path of data in a constellation is optimized according to an actual link state, data transmission time delay is effectively reduced, and data transmission efficiency is improved. According to the time-space change of the flow demand, communication link resources are scheduled in real time, the utilization efficiency of system resources is optimized, and the communication demand of a high-priority task is guaranteed.

The invention discloses an industrial firewall isolation method of a DCS (Distributed Control System). The objective of the invention is to solve the problem that existing industrial network security protection has defects for a DCS system. According to the method, firstly, DCS architecture analysis and risk assessment are carried out, and protection key points and data flow directions are accurately determined; then, a firewall strategy fitting the characteristics of each area (such as a core control unit, an operation station and an external network interface) is customized, and mechanisms such as multiple data encryption are covered; and then firewall deployment and configuration optimization are carried out to ensure protection continuity and rule timeliness. And abnormal conditions are timely handled through a real-time monitoring and dynamic adjustment mechanism, and system recovery under a security event is guaranteed through an emergency response and backup recovery mechanism. And a regular safety drill and evaluation mechanism is arranged to continuously strengthen the protection capability. According to the method, the network security level of the DCS is comprehensively improved, stable and safe operation of industrial production is guaranteed, and the method has remarkable practical value.

The invention relates to the field of circuit breakers, in particular to a primary and secondary fusion pole-mounted circuit breaker which comprises a primary and secondary fusion circuit breaker body, the primary and secondary fusion circuit breaker body comprises a fixing frame, a bottom bin is installed on the rear portion of the fixing frame, bottom columns are arranged in the bottom bin, and bottom grooves are formed in the bottoms of the bottom columns; one side in the bottom groove is in meshed connection with a tooth column, the tooth column is fixedly connected to the middle of the periphery of a rotating shaft, the rotating shaft is rotationally connected to the middle in the bottom bin, the two ends of the periphery of the rotating shaft are fixedly connected with connecting gears through pneumatic expansion heads, and a first branching part and a second branching part are installed on the two portions in the bottom bin correspondingly; through mutual energy storage of the first tapping part and the second tapping part, automatic energy storage work can be achieved, compared with traditional equipment, use of a reset motor is omitted, line arrangement is reduced, meanwhile, electric field interference to the circuit breaker when the reset motor works is avoided, the occurrence rate of abnormity is reduced, and circuit safety is facilitated.

The invention discloses a remote medical data acquisition method and system based on a 5G network, and the method comprises the steps: collecting 5G network state information in real time, calculating a network state information monitoring index, can dynamically adjust a data transmission strategy, and guarantees the timely and safe transmission of key medical data under a complex network condition; meanwhile, a cache and bandwidth cooperation factor and an intelligent cache strategy factor are introduced, the cache depth and bandwidth allocation are dynamically adjusted according to the network state and the medical data priority, intelligent cooperation management of bandwidth and cache is achieved, and continuity of medical data transmission is guaranteed; the constructed early warning model can detect network abnormity in time and trigger alarm, important data are processed preferentially according to an alarm level adjustment strategy, and the abnormity processing and data recovery capability is improved.

The invention discloses a big data real-time processing system and method, and relates to the technical field of big data processing, and the big data real-time processing system comprises a data partitioning module, a data storage module, a data reading module and a data backup module; the data partitioning module is used for partitioning big data entering a block chain in real time by using a partitioning strategy, extracting keywords, and distinguishing the frequency of data blocks according to historical query records; the data storage module stores data blocks through a storage contract, encrypts identifiers of the data blocks and establishes a mapping relation; the data reading module screens data blocks based on the private key set when a user queries, and determines a sending sequence through keyword matching; the data backup module generates a backup after storage, performs different updating strategies for data blocks with different frequencies, and performs adaptive modification on original data by using a hash value; in combination with the block chain technology, multiple advantages of real-time processing, efficient storage, safe reading, flexible backup and the like are integrated, and a comprehensive and efficient solution is provided for big data processing.

The invention relates to the technical field of reverse osmosis membranes, in particular to a polyamide reverse osmosis membrane for high-salinity wastewater treatment and a preparation method of the polyamide reverse osmosis membrane. The invention solves the problem that the polyamide reverse osmosis membrane cannot give consideration to both desalination rate and water flux. The polyamide reverse osmosis membrane comprises a polyester non-woven fabric layer, a polysulfone porous layer, a transition layer and a polyamide layer. The polysulfone porous layer is prepared, and the modified pore-forming agent is added to improve the water flux of the reverse osmosis membrane; the hydrophilicity of the reverse osmosis membrane is improved by preparing the transition layer and adding a hydrophilic auxiliary agent when the polyamide layer is prepared; the flux attenuation rate of the polyamide reverse osmosis membrane is reduced by preparing the polyamide layer; the desalination rate of the reverse osmosis membrane is improved by introducing a composite ion complexing agent and graphene quantum dots into the polyamide layer.

The invention discloses a low-illumination image enhancement algorithm based on deep convolution, which is suitable for the field of image processing, and the core of the algorithm lies in an efficient noise reduction process and accurate decomposition of an illumination image and a reflection image. Comprising the following steps that firstly, a channel noise reduction module works cooperatively through multiple noise reduction strategies, noise in an image is effectively removed, and a foundation is laid for follow-up processing; then, the illumination image decomposition network and the reflection image decomposition network accurately separate the brightness information, texture and color information of the image based on a deep learning structure. And finally, by ingeniously fusing the illumination image and the reflection image, reconstructing a high-quality enhanced image, remarkably improving the brightness, contrast and details of the image, and meanwhile, retaining the natural color of the image. The algorithm has the advantages of being high in noise reduction capacity, accurate in illumination image decomposition, efficient in reflection image decomposition, good in enhancement effect and the like, and an effective and practical solution is provided for image enhancement under the low-illumination condition.

The present invention relates to the technical field of ecological concrete slurry, and specifically to an ecological seed-carrying slurry for the slope of a mine dump, a preparation method and its application. Aiming at the special environmental characteristics of the Inner Mongolia mining area, the present invention prepares a new type of artificial aggregate containing growth domains and seeds, which is composed of a three-layer structure, each layer having a specific function, so as to achieve the purpose of promoting the germination of multi-source species, plant growth, providing nutrition and improving the soil environment. The slurry can not only effectively utilize solid waste and reduce costs, but also adopts in-situ point-type layered hole sowing to shorten the vegetation restoration cycle in the ecological restoration of the slope, greatly improve the survival rate of tree and shrub seed sources, reduce the number of watering times, save restoration costs, and reduce the workload of management. While effectively promoting vegetation restoration, it strengthens the stability of the slope, prevents soil and water loss, slows down runoff and consolidates the soil, providing an efficient and sustainable solution for the ecological restoration of green mines in Inner Mongolia.

The invention relates to a molecular marker which is located at the 47023718th basic group from the 5tail end on a No.16 chromosome of a pig reference genome Sscrofa11.1 and is related to a pig internal farrowing litter weight standard deviation. The invention describes a primer pair and a kit for detecting the molecular marker, and application of the primer pair and the kit in identifying the size of the standard deviation of the litter weight in the pig, screening pig varieties or genetic breeding of the pig. The pig internal litter weight standard deviation of the AA genotype individual is 2.79 kg, the pig internal litter weight standard deviation of the GG genotype individual is 4.38 kg, and the AA genotype individual is obviously smaller than the GG type individual, namely, the pig resilience of the AA genotype individual is higher than that of the GG individual. The AA type individuals are selected, and the pigs with small standard deviation of litter weight and high restoring force in the pigs can be obtained. According to the method, the problem of determining the pig resilience in actual production is effectively solved, the breeding cost is reduced, the pig internal litter weight standard deviation is effectively reduced, so that the pig resilience is improved, the method is high in accuracy and low in detection cost, automatic detection can be realized, and the method has very high practical application value in the aspect of pig breeding.

The invention relates to the technical field of emergency department medical rescue, in particular to cardio-pulmonary resuscitation equipment for emergency department medical rescue, which comprises a back plate, a support frame is fixedly mounted on the left side wall of the back plate, a bearing plate is fixedly mounted at the top end of the support frame, and a pressing and clinging assembly is arranged on the left side of the lower end surface of the bearing plate; the pressing fit assembly comprises an electric push rod fixed to the left side of the lower end face of the bearing plate, a fit pressing plate is fixedly connected to the bottom end of the electric push rod, a storage cavity is formed in the inner bottom end of the fit pressing plate, and a bag cushion is fixedly installed on the bottom end face of the fit pressing plate; a first hydraulic conveying pipe communicating with the interior of the storage cavity is fixedly installed on one side of the upper end face of the attaching pressing plate. The chest of a patient can be tightly fixed through synchronous expansion of the bag cushion, it can be ensured that a pressing head of the cardio-pulmonary resuscitation machine accurately and stably acts on a pressing part of the patient, and pressing position deviation caused by body movement or posture change of the patient is avoided.

The invention provides a multi-level monitoring system suitable for a submarine optical cable. The multi-level monitoring system comprises a central monitoring system, a data acquisition terminal, an underwater robot, sonar equipment and an optical fiber sensor, wherein the central monitoring system comprises a master control platform and a power supply system, the data acquisition terminals are arranged at the starting point and the ending point of an optical cable line, the data acquisition terminals are connected with the central monitoring system in a high-broadband communication mode, and the data acquisition terminals receive data from the optical fiber sensors and send the data to the master control platform. The multiple optical fiber sensors are arranged on the whole optical cable line, the multiple sonar devices are arranged on the optical cable line, scanning data are transmitted to the central monitoring system at regular intervals, and the underwater robot conducts routing inspection along the line at regular intervals through a self-set routing inspection program. According to the invention, the safety and reliability of the optical cable system are obviously improved, the preventive maintenance process is optimized, the environmental adaptability is enhanced, the operation efficiency is improved, the operation cost is reduced, and the data safety is effectively guaranteed.

The invention provides a program crash processing method and device, equipment and a medium. The method comprises the steps that a protection engine is called; when an application program is started, the protection engine is initialized to monitor and capture crash information in the application program, and the crash information comprises abnormal stack information of flash quit of the application program; if the crash information of any application program is captured, processing of the crash information is interrupted according to the type of the abnormal stack information, so that the application program runs normally; the crash information of the application program is collected to generate the crash log, the crash log is uploaded to the server, transmission of the crash information is interrupted, normal operation of the application program is ensured, stability and reliability of the application program are enhanced, the flash quit problem caused by improper operation is reduced, and meanwhile the service life of the application program is prolonged. The debugging and repairing efficiency is improved; the flash quit of the application program is avoided, and the service recovery capability is also improved.

The invention provides an image restoration method based on a memristor array genetic algorithm, and relates to the technical field of image processing. The method comprises the following steps: segmenting an image, converting a data matrix, and storing image gray value information by using a memristor array; encoding and initializing a population; calculating a fitness value, the process degrading the restored image into a degraded image, and comparing the degraded image with the original degraded image; biological heredity is simulated, in the process, the gray value of the image stored by the memristor is selected, crossed and varied, so that the gray value is converged to an optimal solution; and recovering the image, realizing memristive array iteration updating individuals through a genetic algorithm until a stopping criterion is met, and recovering the image. The method has the advantages that the advantages of storage and calculation integration, high-density array, parallel operation and the like of the memristor are utilized, partitioning processing is carried out on the image, the efficiency and the recovery effect of the algorithm are effectively improved, and the problems that due to the fact that the number of image pixels is large, an initial population is difficult to generate excellent individuals, and the convergence speed is low are solved.

The invention provides a space-time signal isolation protection method and system, and relates to the technical field of information security, and the method comprises the steps: 1, constructing a multi-layer time reference architecture in a preset key system, and determining an internal time reference; 2, monitoring and collecting internal signals of a preset type of a preset key system, and establishing a reference feature model; 3, external signals are monitored and collected based on preset monitoring equipment, the external signals are classified based on the internal time reference and the reference feature model, and then abnormal signals are determined; 4, multi-stage isolation protection is started for the classified abnormal signals, and meanwhile, multi-layer early warning and corresponding processing are carried out according to the classified abnormal signals; and step 5, detecting whether the external signal is recovered to be stable, if so, gradually exiting the isolation mechanism through a progressive optimization mode, and introducing the external signal to carry out time synchronization correction. And the space-time protection and recovery capability of the system is enhanced.

The invention provides an arid valley collapse body vegetation restoration method and device, and belongs to the technical field of ecological governance, the arid valley collapse body vegetation restoration method comprises the following steps: obtaining environmental data of a collapse body, determining candidate plant species based on the data, and generating a plurality of candidate restoration schemes; plant combinations of different ages are covered, long-term ecological benefits are ensured, water resource changes are dynamically simulated through a water balance model, water resource data in a collapsed body can be updated in real time, the long-term water resource changes can be predicted, and the adaptability index is determined in combination with water demand data of the plants at the current age. According to the method, the performance of the plant in the restoration scheme can be accurately evaluated, and then the most suitable scheme is selected from the plurality of candidate restoration schemes, so that the restoration plant can adapt to the environmental condition of the collapse body, and the long-term stability and the restoration effect of vegetation are improved.

The invention relates to the field of vehicle diagnosis, in particular to a vehicle fault recovery method, electronic equipment and a medium. The vehicle fault recovery method comprises the following steps: acquiring first data; the first data comprises ECU parameter real-time values and second data; the second data comprises at least one of an ECU parameter preset safety threshold value, an ECU normal operation model and historical fault data; when it is determined that a fault exists according to the first data, a fault recovery strategy is controlled to be executed according to the fault type and the fault reason; and under the condition that it is determined that no fault exists according to the first data, the fault occurrence possibility is determined according to the ECU parameter real-time value and a fault prediction model. According to the method, the robustness can be improved, the adaptability and the recovery capability are enhanced, and the vehicle can still run safely and reliably in a complex environment.

The invention discloses a lightweight social platform image restoration method and system based on dual-branch frequency and space fusion, and a readable storage medium. Comprising the following steps: constructing and training a lightweight social platform image restoration network model based on dual-branch frequency and space fusion; and inputting the low-quality image into a lightweight social platform image restoration network model based on double-branch frequency and space fusion for image restoration to obtain a high-quality image. According to the method, the image definition is remarkably improved, the training process is stabilized, and the robustness and the high efficiency of the model are enhanced.

The invention belongs to the field of power distribution networks, and particularly relates to a power supply recovery method, device and equipment based on an energy storage type intelligent soft switch, and a storage medium, and the method comprises the steps: obtaining the load data of each node of a power distribution network and the power data between the nodes, and enabling the load data to comprise the active load demands and value coefficients of the corresponding nodes at the current moment; the electric data comprises a current value and a resistance value of a line between different nodes at the current moment; inputting the load data and the power data into a pre-established power supply recovery model, wherein a target function of the power supply recovery model is established by taking the minimum loss degree after the power distribution network fails as a target; and performing power transmission on the power distribution network according to output of the power supply recovery model. In the fault stage of the power distribution network, power transmission can be performed on the power distribution network through output of the power supply recovery model, efficient load transfer is realized, the recovery capability of the load of the power distribution network under the fault condition is improved, and important load power supply is guaranteed.

The invention relates to the field of telecommunication, in particular to a distributed acquisition and control networking method. The method comprises the following steps: optical distributed acquisition and control host equipment is in communication connection to form an optical distributed acquisition and control host core looped network; the two optical distributed acquisition and control host devices control one slave optical fiber hand-in-hand connection device to perform management and communication dual-machine redundancy between the two optical distributed acquisition and control host devices, and the two optical distributed acquisition and control host devices form an optical distributed control looped network; the slave optical fiber hand-in-hand connected equipment module is linked with the detection equipment, and the data acquisition efficiency of the optical distributed control ring network is calculated; judging whether one of the optical distributed acquisition and control host equipment is in a disconnected state or not; and calculating the time delay of the slave optical fiber hand-in-hand connection equipment communication connection detection equipment. The problem that monitoring and control are directly influenced by communication interruption among equipment in a part of areas in optical fiber communication is solved; and after the optical distribution type acquisition and control host equipment breaks down, the whole networking cannot work if the dual-redundancy backup is lacked, and the like.

The utility model discloses a modular bus duct structure resistant to lateral impact, which relates to the technical field of bus ducts and comprises two side support plates arranged in parallel, transverse connecting plates are arranged at the top end and the bottom end between the two side support plates, and the two transverse connecting plates are bent back and forth along the horizontal direction. The two transverse connecting plates are arranged oppositely, the top ends and the bottom ends of the conductive copper bars can be clamped at the bent positions formed by the two transverse connecting plates respectively, a gap is reserved between every two conductive copper bars while the transverse connecting plates are kept perpendicular, air can enter the conductive copper bars conveniently, heat dissipation of the conductive copper bars is facilitated, heat accumulation is reduced, and the service life of the conductive copper bars is prolonged. The insulating pads with the trapezoidal grooves can enhance the insulativity of the conductive copper bar on one hand, and the trapezoidal grooves are narrowed towards one side far away from the conductive copper bar, so that the fixing stability of the conductive copper bar can be enhanced, and the conductive copper bar is firmly clamped in the two vertically corresponding insulating pads.

The utility model relates to the technical field of corrugated paper boards, and discloses a pressure-resistant bending-resistant high-strength corrugated paper board, which comprises an upper-layer paper board, a middle-layer paper board, a lower-layer paper board and a corrugated paper core, and the upper-layer paper board and the lower-layer paper board are respectively positioned on the upper side and the lower side of the middle-layer paper board. The corrugated paper cores are arranged between the upper-layer paper board and the middle-layer paper board and between the middle-layer paper board and the lower-layer paper board; a plurality of pressure-resistant elastic strips which are uniformly distributed are arranged in wave gaps of the corrugated paper core; and anti-bending reinforcing ribs are arranged in the upper-layer paperboard and the lower-layer paperboard. The pressure-resistant elastic strips and the bending-resistant reinforcing ribs are additionally arranged in the corrugated board, so that the pressure-resistant and bending-resistant performance of the corrugated board is improved, the thickness of the corrugated board is not greatly changed, and the space utilization rate and the practicability are improved.

The invention discloses a pressure-adjustable constant-temperature wound irrigator, and relates to the technical field of medical instruments, the pressure-adjustable constant-temperature wound irrigator comprises an irrigator body and a liquid storage tank, a liquid inlet cavity is formed in the irrigator body, and a plurality of heating pipes are fixedly mounted in the irrigator body and located in the liquid inlet cavity; the device has the beneficial effects that different pressing plates are pressed by pedaling to perform size adjustment operation, a first rack transversely moves to drive a mounting inner rod to move, then a gear and the rack are matched to drive a valve plate to rotate, and a water temperature detector is fixedly mounted on one side of the flusher body; the pressure of liquid conveying in the liquid outlet pipe is adjusted through rotation of the valve plate, then the effect of adjusting the pressure of output liquid is achieved by pressing the pressing plate, and the hydraulic pressure can be adjusted more conveniently according to the use requirement of a user by arranging the two arc gears capable of enabling the mounting inner rod to move oppositely.

The invention provides a hydrocolloid wrinkle patch, belongs to the technical field of medical instruments and personal care products, and aims at solving the problems that an existing wrinkle patch is uncomfortable to patch and long in service time. Due to the skin-friendly property of the hydrocolloid and the waterproof and breathable properties of the polymer semi-permeable membrane, the use time of the wrinkle patch is longer, the hydrocolloid wrinkle patch can contain beauty components or medical beauty components, and the use experience of a user is improved.

The invention relates to a fan model high and low voltage ride-through process parameter identification method and system based on a Harlisia eagle algorithm and a medium, and the method comprises the steps: firstly verifying a PSASP electromechanical transient simulation model of a direct-drive permanent magnet synchronous wind generating set, and researching the influence of high and low voltage ride-through faults of a direct-drive fan on a power grid; and meanwhile, the influence of key parameters of the direct-driven permanent magnet synchronous wind generating set model on high and low voltage ride-through characteristics is analyzed. According to a fan type test result, wind turbine generator fault transient state control parameter identification is realized by using a Harlisia eagle optimization algorithm, wind turbine generator controller key parameter identification based on PSASP and influence analysis on high and low voltage ride-through characteristics are completed, the deviation between a simulation result and an actual measurement result is within an allowable range, and the method has the advantages of high simulation accuracy and high reliability. Model verification requirements are met, and stable operation of the wind turbine generator is guaranteed.

The invention relates to the technical field of medical apparatus and instruments, and discloses an adjustable orthopedic external fixation support, comprising: a steel needle for connecting with a skeleton and fixing and limiting the skeleton; the fixing mechanism is arranged on the upper side of the steel needle and used for integrally installing and fixing the device on the outer side of a human body needing to be repaired; the clamping mechanism is arranged on the outer side of the steel needle, is connected with the fixing mechanism and is used for clamping the mounted steel needle, keeping the steel needle in a horizontal state and ensuring the stability of the device; the adjusting mechanism is arranged in the middle of the fixing mechanism and used for adjusting the position of the clamping mechanism. Through the height-adjustable clamping mechanism, medical staff can accurately adjust the position of the steel needle according to the fracture condition of a patient, the treatment accuracy is improved, the connecting and separating process of the steel needle is simplified, the use convenience of the device is improved, meanwhile, correct connection of bones is kept, and the recovery effect and speed are accelerated and optimized.

The present application discloses a detachable mobile energy storage device management system, method, equipment and medium, which mainly relates to the field of energy storage device application technology, and is used to solve the existing mobile energy storage system that contains only one energy storage module, which affects efficiency, disconnects the node at the fault, has poor power supply continuity and low distributed power supply utilization efficiency, and the mobile route of the existing mobile energy storage system has an unbalanced power supply problem. It includes: an island division component to obtain a number of power grid islands; a route division component to obtain the calculated difference of distributed power sources; determine the priority of distributed power sources according to weights and calculated differences; complete the connection between the detachable energy storage component and the distributed power source; obtain the calculated difference corresponding to the distributed power source in real time, and when it is less than the preset difference threshold, the distributed power source is disconnected from the detachable energy storage component until an extreme fault elimination signal is received, or the calculated differences of the distributed power sources are all less than the preset difference threshold.

The invention relates to a data encryption transmission method and device with an early warning mechanism, and aims to improve the security and reliability of data transmission. According to the method, a plurality of encryption schemes with unique names are preset and stored at the front end and the rear end of a data transmission system. The front end randomly selects a set of encryption scheme according to security requirements, encrypts request data, sends the encrypted data and a unique name corresponding to the encrypted data to the rear end, and records necessary transmission indexes at the same time. And after receiving the data, the rear end performs decryption and re-encryption processing by using the same encryption scheme, and then sends a result back to the front end. And after receiving the data, the front end decrypts the data and verifies whether the data meets a preset transmission index. And if exceptions are found, such as decryption failure or data delay, the front end records and feeds back the exceptions to the rear end and stops data processing. According to the method, through random selection of an encryption scheme and a bidirectional verification mechanism, the security monitoring and exception handling capabilities in a data transmission process are enhanced.

The invention relates to the technical field of computers, and discloses an operation safety monitoring method and device based on heterogeneous dual systems, equipment and a medium. After entering a working state based on the starting running instruction, guiding other CPU cores except the first CPU core in the BMC to run a service operating system respectively; verifying the service operation system; and when the verification fails, terminating the operation of the service operation system. And the business operation system is verified through the verification operation system, so that the network security of the BMC is improved. And deeper security check is carried out on the service operating system to ensure that the service operating system is not influenced by malicious software or attacks, so that the risk that the server is attacked by the network is reduced. By adopting the new BMC architecture and security mechanism, the security, reliability and performance of the server can be improved, so that the server is more competitive in the market.

According to the multi-modal zero-sample defect detection method based on the search engine and human guidance, the exogenous background image is automatically acquired through the search engine based on the keyword, and the acquisition process of the background image is greatly simplified; meanwhile, a textual feature extractor is enhanced by introducing a texture element network, so that the textual feature extractor can be combined with texture information in an exogenous image, accurate segmentation of the defect is realized when an unseen material or an unseen defect category is processed, and the limitation of the prior art in facing rare defects is overcome; that is to say, the problem that a traditional defect detection algorithm depends on a large-scale data set is solved by introducing a multi-modal zero sample learning method and combining feature alignment of the text and the exogenous texture image.

The invention relates to the technical field of artificial intelligence, in particular to an efficiency evaluation method and device for intelligent operation and maintenance of a cluster system. According to the method, the efficiency of the cluster system in different operation stages can be scientifically and accurately evaluated by constructing the reasonable efficiency index system, the network average efficiency calculation model, the task execution balance degree evaluation model and the elasticity evaluation model. According to the method, the deep reinforcement learning algorithm is introduced, the multi-dimensional situation information is collected, analyzed and predicted in real time, and the intelligent operation and maintenance decision-making efficiency of the cluster system can be effectively improved.