87results about How to "Realize failure prediction" patented technology

The invention relates to a battery management system. The battery management system comprises a monitoring device, a mobile terminal and a cloud processor, wherein the monitoring device is used for detecting and transmitting working parameters of each single battery in real time when the battery is charged or discharged, comparing the working parameters of the single battery with the standard parameters, automatically adjusting the working state of the single battery when the working parameter exceeds the standard parameters and transmitting alarming information; the mobile terminal receives and displays the alarming information; the cloud processor receives and stores the working parameters. The invention aims at providing the battery management system, the voltage and temperature of the single battery are monitored and recorded in real time, the working state of a battery pack and the charging-discharging times are detected, and the integral temperature of the battery pack is detected, so that the working state of a charging-discharging device and a heating device is controlled, the functions such as battery quantity management, temperature management, fault pre-judgment and diagnosis and the like can be realized, and the safety and reliability of the battery can be guaranteed; a great amount of data is stored in a cloud server and can be pushed to enterprises for producing the battery or he battery management system users.

The invention discloses a distribution network fault prediction method and system based on big data, and belongs to the technical field of power network safety maintenance. The method comprises the following steps of: S1, acquiring original data, S2, processing the original data by using Tableau software so as to obtain analyzable sample data, wherein the processing comprises data cleaning, data transformation and data integration; S3, analyzing the correlation of the fault influencing factors of the distribution network; S4, building a fault prediction model based on an improved random forestalgorithm; wherein the improved random forest algorithm generates a plurality of classification trees which are combined through modes of voting or arithmetic mean decision. The method achieves the fault prediction of the local distribution network line. And fault early warning information is timely issued to provide targeted line operation and maintenance guidance for power distribution networkoperation and maintenance departments. The hidden trouble of line failure is eliminated in advance, the failure rate of the distribution network is reduced, and the power supply reliability of the distribution network is improved.

The invention discloses a health evaluation index and failure predication method for a DC (Direct Current)-DC convertor. The method comprises the following steps of: collecting state signals of input voltage, input current, output voltage and output current of a DC-DC circuit, calculating the average values (Uin, Iin, Uout, Iout) of the signals, as well as the equivalent load impedance modulus value |Z| and power consumption equivalent resistance (Rloss) of the circuit; and establishing a relation model of Uin, |Z| and Rloss by utilizing a fractional order nerve network; then calculating the equivalent load impedance modulus value, which is utilized as model input, of a health circuit under the conditions of rated input voltage and rated output power, acquiring the corresponding Rloss used as a circuit health evaluation index, calculating the difference value between the circuit health evaluation index and a reference value, and determining the circuit health index according to the difference value and a set change threshold so as to realize the circuit health evaluation; and finally carrying out time sequence predication on the circuit health evaluation index, acquiring a health evaluation index of the next moment, and predicating the health condition of the circuit. According to the method disclosed by the invention, the power consumption equivalent resistance under the conditions of rated input voltage and rated load is utilized as the circuit health evaluation evaluation index, the influence of the input voltage and load changes on the power consumption equivalent resistance is avoided, and thus the health evaluation and the failure predication of the circuit are accurately realized.

The invention provides a mechanical equipment fault diagnosis method based on a machine learning classification algorithm. The mechanical equipment fault diagnosis method comprises the following stepsthat step 1, an acceleration sensor is used for collecting vibration signals of key points of mechanical equipment, and original waveforms of the vibration signals are stored; step 2, the collected vibration signals in the step 1 are screened and judged; step 3, the screened vibration signals are preprocessed; step 4, acceleration signals, velocity signals and envelope signals obtained in the step 3 are subjected to feature extraction; and step 5, the obtained feature vector in the step 4 is input into a fault classification model, and the model outputs fault diagnosis results corresponding to the equipment. The mechanical equipment fault diagnosis method based on the machine learning classification algorithm establishes an intelligent diagnosis model of the mechanical equipment, and intelligent diagnosis of fault of the mechanical equipment is further realized.

The invention provides a vehicle state analysis method, a vehicle state analysis system and a ground comprehensive information analysis subsystem. The vehicle state analysis method comprises the following steps that data sent by an information terminal and data sent by a ground collector are received; the received data is analyzed, and service fields are extracted from the analyzed data; and according to the extracted service fields and historical data, calculation and analysis are carried out through a mathematical operation model to obtain and display the current and subsequent working states of a vehicle. According to the vehicle state analysis method, the vehicle state analysis system and the ground comprehensive information analysis subsystem, more comprehensive data of the vehicle can be obtained, calculation and analysis can be carried out through the mathematical operation model according to the obtained data, and analysis results of the current and subsequent working states ofthe vehicle can be obtained, so that failure conditions of the subsequent work of the vehicle is predicted, for example, the time and the reason of the future failure of the vehicle can be predicted,failure prediction can be realized, and further maintenance can be carried out as early as possible before the failure of the vehicle occurs.

The invention discloses an aero-engine turbine blade reliability digital twinning modeling method, relates to the technical field of aero-engines, and aims to realize high fidelity simulation and tracking of aero-engine turbine blade damage and life loss and quickly and efficiently evaluate the reliability and residual life of the aero-engine turbine blade. According to the invention, the turbineblade is taken as an object, aiming at the characteristics of multi-source and multi-time-scale use reliability related information, few fault samples and the like of the turbine blade, a physical model driving and data driving fusion method is adopted, that is, the specialty of a physical model in the aspect of explaining specific data is fully utilized, and a turbine blade digital entity model based on a high-precision physical model and multi-source data is constructed by means of the advantages of a machine learning method in the aspects of dimension reduction and multi-modal data fitting.The model can reflect the physical characteristics of the turbine blade and the variable characteristics dealing with different environments and damages, and can realize the product operation reliability dynamic evaluation and residual life prediction of multimode information fusion.

The invention provides a CNN-based power equipment fault judgment and early warning method, a terminal and a readable storage medium. The method comprises the steps of obtaining test data; preprocessing the data; processing the data by using an offline model; and performing fault prediction of data. According to the method, the coal mill data is modeled through a deep learning method, fault prediction is achieved, mass historical data of coal mill equipment are fully mined through an existing data mining and machine learning modeling method, and an efficient and practical model is establishedto conduct detection and early warning on the real-time state of the coal mill. Knowledge and experience of experts and operating personnel are combined with data mining and machine learning methods and complemented with each other. The data can be automatically analyzed and modeled according to the data characteristics, and the threshold of operating personnel is lowered. The fault prediction model of the coal mill established by the invention can contain more complex causal relationships implicit among the indexes, so that the possibility of loss of a large amount of effective information isavoided, and the result is relatively reasonable and accurate.

The invention relates to a method for predicating an underground fault of a sucker-rod pump oil pumping well on the basis of a multivariable grey model. The method comprises the following steps of: collecting a ground indicating diagram by an indicating diagram remote data collecting system; converting the ground indicating diagram into an underground pump indicating diagram; extracting a curve moment feature vector capable of representing the graphic feature of the underground pump indicating diagram according to the curve moment theory; predicating each feature vector by the multivariable grey model; and calculating the grey correlation degree of each fault type in predicating samples and standards sets, wherein the fault type corresponding to the maximum grey correlation degree is the fault type of the predicating samples. The working system of the oil pumping well can be reasonably regulated, so that the production efficiency of the oil pumping well is improved, and the fault predication on the underground work condition of the sucker-rod pump oil pumping well can be realized; the calculation is simple, the required samples are few, and the model is applicable to data sets in any distribution; and the prediction accuracy is high, and particularly, the prediction precision on the gradually varied fault type is high.

The invention discloses a method for producing formula food for special medical purposes by using a manufacturing execution system. The method comprises the following steps: generating a sales order in an ERP system; making a production scheduling plan according to the sales order; transmitting a production scheduling plan in the ERP to the MES system by using an enterprise connector; wherein theenterprise connector is an API interface, a message bus or an intermediate table; and acquiring data; centralized acquisition and control of production system data and environment system data are realized through an SCADA system platform, real-time acquisition and storage of massive process data are realized, and data support is provided for an MES system; the production efficiency is improved, and the production value energy consumption is reduced; through customized production process flow control and information management, material and energy loss caused by misoperation of personnel and unreasonable production is reduced, and the product productivity is maximized.

The invention discloses a phase-shifting full-bridge convertor real-time fault diagnosis method and a system; two lower bridge arms of a phase-shifting full-bridge convertor are provided with current sensors, in a positive half of a complete switching cycle of the phase-shifting full-bridge convertor, the sensor of the front lower bridge arm obtains a power tube anti-parallel diode freewheel current signal; in the negative half of the switching cycle, the sensor of a rear lower bridge arm obtains a power tube anti-parallel diode freewheel current signal; and the faults of the phase-shifting full-bridge convertor are determined based on the logic relationship between the power tube anti-parallel diode freewheel current signal and a power tube driving signal by integrating the power tube driving signal with the freewheel current signal obtained by the sampling of the positive half or the negative half. The invention can quickly and accurately position the faults and determine the fault types, establish a fault database, and realize fault prediction.

A method for predicting mechanical seal leakage failure of in-service rotating equipment includes collecting and sorting leakage failure data, including associated failure and non-associated failure;establishing a data-driven model of mechanical seal leakage fault, including establishing normal distribution model, logarithmic normal distribution model, exponential distribution model and two-parameter Weibull distribution model based on the running fault data of mechanical seal of in-service rotating equipment, and carrying out parameter inversion and parameter verification to obtain a data-driven model of mechanical seal leakage fault; selecting the data-driven model of mechanical seal leakage fault; and according to the data-driven model of mechanical seal leakage fault, predicting the reliability life of mechanical seal of in-service rotating equipment. The invention provides a mechanical seal leakage fault prediction method with wide application range and of considering the actualuse condition of the equipment, which can provide an effective technical means for the long-period operation of the rotating equipment.

The invention relates to the technical field of on-line lightning arrester monitoring, in particular to a non-residual voltage monitoring device for gapless metal oxide lightning arresters and a measurement method. A non-residual voltage method is utilized to monitor the operating state of a lightning arrester on line. The non-residual voltage monitoring device comprises a case and an internal circuit board; the case is provided with an outer cover, which is provided with a display screen window, and an LED (light-emitting diode) display is mounted on the end cover and displays in the window; the outer cover is also provided with lugs, cable holes and jacks; and the internal circuit board comprises a sensor part, a central processor MCU (microprogrammed control unit), a clock circuit, a memory circuit, a signal measurement and conversion circuit, a protection circuit and a communication circuit. The non-residual voltage monitoring device can monitor the parameters of the lightning arrester, such as total current, resistive current and frequency of action, and has the functions of local display, remote communication, data storage and the like. The non-residual voltage monitoring device is applicable to the on-line monitoring of lightning arresters of various voltage levels, can completely substitute for conventional total current meters and act counters, and has a broad application prospect.

The invention relates to a controller of a converter of a dual-fed wind power generator, which is formed by connecting a touch display screen, a PLC module, a data acquisition and control module and a network interface module, wherein the touch display screen is connected with the PLC module, the PLC module is connected with one end of the network interface module, the other end of the network interface module is connected with a control and pitch control system of a wind turbine, the PLC module is further connected with the data acquisition and control module, and the data acquisition and control module is simultaneously connected with a rotor side power module and a network side power module. The controller adopts the modular design and can realize the double closed-loop control of a power outer loop and a current inner loop through acquisition, analysis and treatment of various analog signals, digital quantity signals and rotation speed signals, and simultaneously realize the functions of centralized acquisition of data, fault prediction, diagnosis and the like, and has the characteristics of strong anti-interference ability, abundant network interfaces, fast measurement speed, high control precision, friendly man-machine interface and the like.

The invention discloses a large-scale equipment online monitoring and fault prediction system. The system comprises a data acquisition assembly, a field control end, a central control end and an equipment monitoring end; the data acquisition assembly acquires one or more feature data of the large-scale equipment in real time; the central control end traces and analyzes historical data of the large-scale equipment and constructs a fault prediction model; the invention includes inputting the feature data received in real time into the fault prediction model, performing feature extraction, identification and classified learning on the feature data of the large-scale equipment by using an artificial intelligence algorithm including a convolutional neural network, and performing fault prediction and diagnosis on the large-scale equipment; and the equipment monitoring end receives and displays the characteristic data of the large-scale equipment sent by the field control end in real time, and receives and displays the fault prediction and diagnosis result of the central control end at the same time. According to the invention, fault prediction and timely maintenance of the large-scale equipment are effectively realized, the purpose of nipping in advance is achieved, and the large-scale equipment is ensured to be in a safe and reliable operation state.

The embodiment of the invention discloses a method and a device for processing the whole life cycle state data of a railway freight car. the accumulated mileage of the train running from the time of repairing to the time of failure is counted, and isassociated with the component information of each component, for example, the cumulative mileage is associated with the repair unit that repairs the component, so that each repair unit can be evaluated and the mileage that can be driven after repairing by a repair unit can be predicted. This method integrates the information of train failure and mileage, and can be used to analyze the relationship between the operation time of railway freight car accessories, mileage and maintenance unit.

The invention discloses a wireless control system for an electric bicycle. The wireless control system for the electric bicycle comprises an electric bicycle controller, a Bluetooth module and a terminal service module, wherein the controller is connected with the terminal service module through the Bluetooth module; and the controller comprises a main control chip, a power supply module, a serial communication module, an overcurrent protection circuit, a current detecting circuit and a handle signal detecting circuit. The invention further discloses an electric bicycle using the wireless control system. According to the invention, Bluetooth technology is integrated with the electric bicycle controller; the controller is wirelessly connected with an intelligent terminal through Bluetooth; operation parameters, state parameters and product fault situations of the electric bicycle are displayed on APP; and the gear of the electric bicycle can be adjusted through changing gear information on the APP. According to the invention, the intelligent level and riding safety of the electric bicycle are enhanced, bicycle faults can be predicated to increase the maintenance efficiency of the electric bicycle, user loss is reduced, the user waiting time is also reduced, and the after-sale service is optimized.

The invention relates to an ocean platform vibration noise comfort monitoring device and an online monitoring method. The device comprises a data storage and analysis platform which is connected witha plurality of distributed test terminals through a switch; the input end of each distributed test terminal is connected with a sensor; a computer terminal is further included and connected with the data storage and analysis platform through the switch; the distributed test terminals collect data transmitted by sensors, convert time domain signals into frequency spectrum data and transmits the frequency spectrum data to the data storage and analysis platform; the data storage and analysis platform compares the frequency spectrum data with a preset limit value and historical data, and when thefrequency spectrum data exceeds the limit value, alarm is raised. According to the invention, vibration of equipment can cabins in an ocean platform as well as air noise are monitored in real time, sothat a vibration noise abnormity fault can be eliminated timely, and it is ensured that the platform meets the comfort requirement of ship-level symbols.

The invention relates to the field of maintenance scheduling, in particular to a train air conditioner maintenance scheduling system and a working method thereof, and the train air conditioner maintenance scheduling system comprises a fault prediction server, a data screening module and a data transmission module, wherein the data screening module is suitable for sending screened train air conditioner real-time operation data to the fault prediction server through the data transmission module; and the fault prediction server is suitable for predicting an air conditioner fault according to thetrain air conditioner real-time operation data. And fault prediction of the train air conditioner is realized.

The invention discloses a fault prediction method for a photovoltaic inverter. The fault prediction method comprises the following steps that historical monitoring signals of a photovoltaic inverter cluster of the same photovoltaic power station serve as an original feature library; a main feature matrix of the photovoltaic inverter cluster at each sampling time is extracted from the original feature library through a sparse self-encoding algorithm; a cluster center photovoltaic inverter at each sampling time is searched based on a fast clustering algorithm; a cumulative eccentric distance matrix of the photovoltaic inverter cluster is calculated; and the cumulative eccentric distance matrix is subjected to normalization processing, an early warning threshold value is set, and finally, prediction of the fault of the photovoltaic inverter is achieved. According to the fault prediction method, prediction of the fault of the photovoltaic inverter is achieved, on-line operation can be conducted, calculation is convenient, special requirement limitation is avoided, the fault prediction method is suitable for photovoltaic inverter clusters with different scales, portability is good, overhaul personnel can establish a reasonable and effective maintenance plan advantageously, and safe and stable operation of a microgrid is ensured.

The present invention discloses a conveyor belt fault prediction method, system and device based on the HTM (Hierarchical Temporal Memory). The method comprises the steps of: collecting state data ofa conveyor belt device; and inputting the state data of the conveyor belt device into an HTM algorithm model for processing so as to output abnormal scores. The system comprises a collection unit anda processing unit. The device comprises at least one storage and at least one processor configured to load a program in the storage to execute the conveyor belt fault prediction method based on the HTM. The conveyor belt fault prediction scheme does not need manual operation to define a rule in advance, does not need manual operation to perform rule association mining between states of a lot of states and does not need manual operation of updating the rule to greatly reduce the workload and work burden of operators, improve the working efficiency and operation usage convenience and greatly improve the fault prediction accuracy and real-time effectiveness. The conveyor belt fault prediction method, system and device based on the HTM can be widely applied in the airport transmission facilities.

The invention discloses a variable-speed variable-load large rolling bearing fault prediction system and method. The system comprises a variable-speed variable-load rolling bearing experiment table and a bearing intelligent information system. The variable-speed variable-load rolling bearing experiment table is used for simulating the variable-speed variable-load working environment of the rollingbearing and comprises a control system, a loading device and a testing system, and the testing system is connected with the control system and the loading device; the bearing intelligent informationsystem carries out state monitoring and fault prediction on the rolling bearing based on data collected by the variable-speed variable-load rolling bearing experiment table and comprises a database, adata processing module, a state monitoring module and a fault prediction module, the database is connected with the state monitoring module, and the database is sequentially connected with the data processing module and the fault prediction module. By constructing the rolling bearing experiment table, the variable-speed and variable-load working characteristics of the bearing are simulated, and the collection of full-life-cycle bearing operation signals is realized; a system training data set is obtained by collecting the operation state information of the bearing, and fault prediction of therolling bearing under variable speed and variable load is achieved.

An energy storage converter fault prediction method includes the following steps: taking the historical monitoring signals of the energy storage converter cluster of the same battery energy storage system as the original feature library; using a sparse self-coding algorithm to extract the main feature matrix of the energy storage converter cluster at each sampling time from the original feature library; using a fast clustering algorithm to search for the central energy storage converter at each sampling time; calculating the cumulative eccentric distance matrix of the energy storage convertercluster, normalizing the cumulative eccentric distance matrix and setting the warning threshold, and finally realizing the fault prediction of the energy storage converter. The invention realizes thefault prediction of the energy storage converter, can be operated on-line, is convenient to calculate, has no special requirement restriction, is suitable for the energy storage converter clusters ofdifferent scales, has good portability, is favorable for the maintenance personnel to establish a reasonable and effective maintenance plan, and ensures the safe and stable operation of the power network.

The invention provides a new energy automobile electric control system fault prediction method based on working condition data. On the basis of collecting fault data, such as data of fields of a controller output deviation, controller response data, an instrument display deviation and the like, and fault samples of a new energy automobile electric control system, the data are used as support of anew energy automobile fault prediction database, the fault samples are used as learning samples, a neural network-based electric control system fault prediction model is established in combination with a neural network prediction method, a final probability prediction matrix is obtained through matrix operation and conversion of a softmax layer, and a fault with the maximum probability is selectedas a final prediction result. A training method of the model is a random gradient descent method, the neural network model which can be used for actual electronic control system fault prediction is finally formed by continuously iterating until the error is less than a threshold or the number of iterations is greater than a set value, and the prediction accuracy rate of the fault is more than 96%.

The invention relates to a health diagnosis method for a diesel generating set; and the method comprises the steps: combining a set health scoring system with a set fault recognition model, carrying the set health scoring system and the set fault recognition model on an intelligent IO controller to obtain a set of complete set health diagnosis system, and obtaining a current state operation healthindex of the set through employing set parameters as input; a parameters change is reflected by a health score in real time when the parameters change, and early warning is carried out in advance when a certain parameter is seriously deteriorated, so that fault prediction and fault diagnosis are realized. In the early stage of a fault or even when the state of the set is slightly abnormal, judgment and warning can be conducted in time, the operation health state of the set is mastered in real time and prevented in advance, sudden accidents are prevented, and the stability of the set and the system safety are improved. And a set of universal fault prediction algorithm solution is established and is gradually developed to provide a whole set of intelligent function development solution forfault diagnosis and fault prediction for various ship mechanical auxiliary equipment.

The invention discloses an ETC door frame operation state monitoring system for an expressway, and the system employs an advanced information processing mechanism to complete the real-time processing of big data, achieves the monitoring of all equipment, achieves the evaluation of road network operation indexes, carries out the real-time notification and prompt of resource alarm information, and achieves the multi-layer and multi-point comprehensive combined application. The system comprises a system operation and maintenance cockpit, a monitoring center, a report center, an equipment management sub-module, an operation and maintenance management sub-module, a system management sub-module and the like. Core functions of the monitoring center comprise software version check, heartbeat state statistics, heartbeat real-time state statistics, transmission delay statistics and system alarm information. Core functions of the report center comprise a whole province toll station transaction amount statistical table, expressway exit station traffic flow statistics, expressway truck 10% discount amount statistics, expressway exit traffic flow statistics, bottom-toll rate transaction amount, transaction amount statistics and the like. All kinds of data used by the system are obtained according to a standard interface protocol and a network management protocol.

The invention provides a control system of indirect adiabatic and evaporative cooling devices of a data center. The control system comprises a controller, a sensor and a management server, wherein thecontroller comprises a plurality of interfaces, the controller is in communication connection with each indirect adiabatic and evaporative cooling device in a machine room of the data center throughat least one connection mode via the interfaces; the sensor is used for acquiring the environment temperature and the environment humidity of the interior of the machine room and the outside of the data center; the controller is used for acquiring operating data, environment temperature and environment humidity of each indirect adiabatic and evaporative cooling device, sending the operating data,the environment temperature and the ambient humidity to the management server, receiving control commands of the indirect adiabatic and evaporative cooling devices from the management server, and controlling the indirect adiabatic and evaporative cooling devices through the control commands; and the management server is in communication connection with the controller, and the management server isused for receiving operating data, environment temperature and environment humidity from the controller, and generating a control command according to the received data and operating states of computing equipment in the machine room.

The invention discloses a source network load control second-level network remote monitoring analysis platform, and the platform comprises a network equipment management system, a data acquisition server, a network monitoring system and an early warning analysis system; the data acquisition server acquires transformer substation operation data received by the monitoring hosts of all transformer substations in the current level and corresponding next-level transformer substation operation data; the network monitoring system monitors and analyzes the operation data of each transformer substationby combining the operation data of the next-stage transformer substation and the configuration data of each network device; and the early warning analysis system performs early warning analysis on front-stage network equipment and rear-stage network equipment of the abnormal network equipment in combination with the hierarchical relationship between the substations. According to the invention, the hierarchical relationship between the substations and the relationship between the network devices contained in the substations are combined to monitor and analyze each network device of all the substations in the area, so the fault monitoring efficiency of the network devices is greatly improved, and the fault prediction of the network devices which may have anomalies is realized.

The invention belongs to the technical field of operation and maintenance monitoring, and particularly relates to a method for detecting the operation state of a sliding door lock of a platform door system. According to the detection method, the time difference between the time when the sliding door lock sends out an electromagnet attraction signal and the time when the sliding door lock receivesa lock in-place switch action signal is calculated, the time difference between the time when the sliding door lock receives a closing in-place switch action signal and the time when the sliding doorlock receives the lock in-place switch action signal is calculated, and the operation state of the sliding door lock in the platform door system is detected through a time calculation method. Detection is easy, accuracy is high, the operation reliability of the sliding door lock can be greatly improved, and the equipment failure rate is reduced.

The invention discloses a digital power station delivery method based on BIM and GIS, a medium and equipment. The method comprises the following steps: determining a digital delivery rule; determining a digital deliverable delivery requirement; determining a digital delivery scheme to establish a BIM, a GIS model or a BIMGIS model of a digital deliverable in the power station; digital delivery information association: determining that a digital deliverable in the power station needs to deliver associated information and associates the associated information to the deliverable; taking the BIM, the GIS model or the BIMGIS model of the deliverable associated with the completed information and the information associated with the deliverable as delivery objects for delivery; and after the delivery is completed, checking the detailed information of the deliverable through the local window. According to the method, paperless digital interaction can be achieved, the technical problems that traditional delivery is low in efficiency and accuracy are solved, a source is provided for digitization of a power station through three-dimensional design, source-to-end direct connection of information is achieved, and it is fundamentally ensured that the information is correct.